International Construction Specialty Conference of the Canadian Society for Civil Engineering (ICSC) (5th : 2015)

Feasibility of plug-load monitoring and energy-saving interventions in residential and office buildings… Kosonen, Heta K.; Kim, Amy Jun 30, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52660-Kosonen_H_et_al_ICSC15_018_Feasibility_Plugload_Monitoring.pdf [ 1.07MB ]
52660-Kosonen_H_et_al_ICSC15_018_Feasibility_Plugload_Monitoring_slides.pdf [ 653.77kB ]
Metadata
JSON: 52660-1.0076491.json
JSON-LD: 52660-1.0076491-ld.json
RDF/XML (Pretty): 52660-1.0076491-rdf.xml
RDF/JSON: 52660-1.0076491-rdf.json
Turtle: 52660-1.0076491-turtle.txt
N-Triples: 52660-1.0076491-rdf-ntriples.txt
Original Record: 52660-1.0076491-source.json
Full Text
52660-1.0076491-fulltext.txt
Citation
52660-1.0076491.ris

Full Text

5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   THE FEASIBILITY OF PLUG-LOAD MONITORING AND ENERGY-SAVING INTERVENTIONS IN RESIDENTIAL AND OFFICE BUILDINGS ON THE UNIVERSITY OF WASHINGTON CAMPUS  Heta K. Kosonen1,3, Amy A. Kim2  1 Department of Civil and Environmental Engineering, University of Washington, USA 2 Department of Civil and Environmental Engineering, University of Washington, USA 3 hetak@uw.edu  Abstract: The University of Washington (UW) is aiming to reduce the overall electricity consumption on campus as part of its Climate Action Plan launched in 2009. To achieve this goal, UW installed 216 smart grid meters and automatic heating, ventilation, and cooling control systems across the entire campus and acquired over 200 sets of plug-load monitoring equipment. The university used the smart grid data and the monitored plug-load data to test how occupants in selected residence halls responded to receiving detailed information about their energy usage patterns, its environmental impacts, and associated costs. The experiment demonstrated that in residence halls, plug-load monitoring did not have any significant impact on the occupants’ electricity consumption. Hence, there is still a need to further assess which strategies are effective in achieving long-term electricity reduction goals for the university. The goal of this study was to conduct a comparative analysis by replicating the plug-load analysis conducted in residence halls in a faculty/staff office setting. The study entailed interviewing university administrators that were involved in the residence hall plug-load study. Interviewees were asked questions about the findings, shortcomings, and recommendations for future studies. Also, this study characterized the load profiles of the faculty/staff offices by monitoring the plug-load consumption in four offices for nine weeks and explored plug-load reduction interventions applicable to office settings. The study found that the unreliable network connection caused frequent disruptions in data collection and strong bias in the individual electricity consumption data. The inventory of electronic appliances in the monitored offices revealed a high variability in the number of devices which lead to variations in base consumption and peak plug loads between faculty offices, and lack of occupant engagement was found to be the main challenge in the implementation of plug load monitoring campaigns. The results provide universities around the country with valuable information and insights on how to design and implement an on campus plug-load reduction intervention with quantifiable energy-saving potential. 1 BACKGROUND Worldwide reductions in greenhouse gas emissions are needed to mitigate the most severe impacts of climate change. Commercial buildings account for roughly 18 percent of the total annual energy use in the United States (US DOE 2014). Therefore, sustainability initiatives at workplaces offer a substantial opportunity to reduce the harmful effects on the environment. The University of Washington (UW) is aiming to reduce the overall electricity consumption on campus as part of its Climate Action Plan launched in 2009. Currently, miscellaneous electricity loads (e.g., computers and office equipment) from offices account for approximately 12 percent of the total electricity 18-1 used in the UW, which means that an average of 36 million kilowatt hours of electricity is used annually on plug loads (University of Washington 2012). According to the results presented in recent literature, a well-designed energy intervention in the UW offices has the potential to save up to $360,000 of UW’s annual electricity costs. Furthermore, reducing plug loads would not only help the university to reach its carbon neutrality goal but would also improve the power quality on the local grid and moderate peak electrical demand in the Seattle area. Traditionally, energy-conservation efforts in office environments have been implemented through technological or operational modifications (Starik & Marcus 2000). However, humans are the main operators of technology, and a failure of the human component can fail the entire energy-efficiency initiative. Thus, instead of focusing solely on technological solutions, recent research has shifted to investigate the effects of occupant behavior on building energy use (Fischer 2008, Azar & Menassa 2010, Masoso & Grobler 2010, Schweiker & Shukuya 2010, Kamilaris et al. 2014). The results of these studies have shown that the energy-saving potential of behavioral change is comparable to, and even higher than, that of technological solutions (Masoso &Grobler 2010, Schweiker & Shukuya 2010). Some estimates even suggest that the occupants control or impact up to 50 percent of a building’s energy use and that changing occupant behavior patterns gives the most effective reductions in energy use (Kamilaris et al. 2014). Changing occupant behavior in offices and other commercial buildings is not without its challenges. A wide variety of studies have looked into the different types of interventions that could most effectively result in electricity savings. One of the most applied measures of impacting and controlling occupant energy use is giving occupants regular feedback on their energy usage patterns (Jain et al. 2012, Jeong et al. 2014, Gulbinas et al. 2014, Hargreaves et al. 2010, Hargreaves et al. 2013, Pereira et al. 2013, Froehlich et al. 2010, Vine et al. 2013). Because feedback frequency and accessibility have been found to correlate positively with the impact on energy reductions (Abrahamse & Steg 2011), most of the recent research on occupant behavior has used real-time monitoring solutions (Jain et al. 2012, Gulbinas et al. 2014, Jain et al. 2013a, Jain et al. 2013b, Ueno et al. 2006). Various studies have shown that frequent feedback is generally effective and correlates negatively with the energy consumption rate (Faruqui et al. 2010, Siero et al. 1996, Vassileva et al. 2012, Murtagh et al. 2013). However, its effects are often temporary, as the engagement of the participants has been repeatedly observed to reduce over time (Hargreaves et al. 2010, Ueno et al. 2006, Murtagh et al. 2013). Furthermore, not everyone is interested in receiving feedback on their electricity consumption: In their study, Murtagh et al. found out that 41% of the participants did not access their individualized feedback even once. These results indicate that in order to design effective electricity interventions with consumption feedback, the focus should be in long-term participant engagement. When implemented successfully, high-frequency electricity feedback can result in total electricity savings of about 20 percent (Murtagh et al. 2013, Acker et al. 2012, Ecova 2011). UW has already installed 216 smart grid meters and automatic heating, ventilation, and cooling control systems across the entire campus and acquired over 200 sets of plug-load monitoring equipment. In 2013, the university used the smart grid data and the monitored plug-load data to test how occupants in selected residence halls responded to receiving detailed information about their energy usage patterns, its environmental impacts, and associated costs. The project team examined electricity use in each of the buildings over a ten-week period in order to understand which intervention, a technology intervention or education intervention, would have a greater effect (if any) on floor-wide energy consumption. The experiment demonstrated that in residence halls, neither educational nor technical plug-load reduction interventions had any significant impact on the occupants’ electricity consumption: Occupants appeared to have higher energy use throughout the study, and educational intervention failed to produce statistically significant results. The research group listed small sample size, inexpensive energy, subjects who do not pay individual energy bills and technical difficulties as some of the factors that may have contributed to these results. Despite the inconclusive results, the team suggests that University has potential to educate students and successfully reduce energy use though other approaches. However, the use of plug load monitoring systems was not recommended as it was found to be an expensive and ineffective tool for changing energy behavior in the context of the University’s residence halls (Black et al. 2014). 018-2 The goal of our study is to conduct a comparative analysis by replicating the plug load monitoring campaign conducted in residence halls in a faculty/staff office setting. In addition to collecting monitoring data, we conducted a survey of the university administrators that were involved in the 2013 residence hall plug-load study to learn more about the challenges and opportunities related to plug load monitoring on campus setting. The study will serve as a preliminary study for a plug load monitoring campaign that will be implemented in one of the office buildings at UW campus later in 2015. We believe that the results will provide universities around the country with valuable information on how to design and implement an on-campus plug-load reduction intervention with quantifiable energy-saving potential. 2 METHODS 2.1 Plug load monitoring  2.1.1 Equipment Plug load monitoring systems with control capability were installed in four faculty offices. The systems consisted of smart power sockets and strips, a Wi-Fi-connected touchscreen monitor with control capability over smart sockets and strips, and an online user account for data collection. (Figure 1) A total number of 20 appliances were plugged into the smart sockets and strips that were connected to the touchscreen monitors over Wi-Fi. High power appliances, such as refrigerators, fans and microwaves had to be excluded from the study as the monitoring system only supports devices with up to a maximum of 15 amps (EnergyHub Inc. 2011). In addition, University’s IT staff requested that desktop computers were kept on at all times to allow for software and security updates.  Figure 1. Set-up of the plug load monitoring system 2.1.2 Installation and education The monitoring systems were set up over a period of one week. Prior to installation, the office occupants conducted an inventory of their electronic appliances together with the research staff and identified the appliances that were to be connected to the monitoring system. In addition, each appliance was given a status on the basis of occupant’s requests: If appliance was given an “always on” status, it would stay on even if the smart strips and sockets were turned off. Other appliances with an “on-off” status would turn off normally when the power to the strips and sockets was cut off.  The installation process consisted of three phases. In first phase, sockets and strips were connected to the touchscreen monitor by using strip- and socket-specific set-up codes. In the second phase, the electronic appliances in each socket and strip were named in order to allow appliance-by-appliance electricity monitoring. In the last phase, the status of each appliance was determined by typing the information in the touchscreen monitor. In order to minimize the disturbance to the occupants, installation work was completed when offices were unoccupied.  018-3 3 RESULTS 3.1 Plug load monitoring 3.1.1 Monitoring system installation & operation Depending on the amount of electronic devices in the office, the installation process took approximately 30 to 45 minutes per office. After installation, the exact locations and identifying information of all smart sockets, strips and touchscreen monitors were collected to a directory, which was kept up to date about the performance of the study equipment through the monitoring period. Apart from couple of malfunctioning smart sockets, the monitoring hardware functioned as expected and did not require maintenance over the 9-week study period. The total time used for hardware installation and operation was approximately 1-1.5 hours per office per 9 weeks. The initial plan was to connect all monitoring systems to the University’s Wi-Fi but after unsuccessful attempts in all office rooms it became evident that the touchscreen monitors did not communicate with the server: users were unable to access their devices remotely and no electricity consumption data were saved to the system database online. The issue was temporarily resolved by connecting the monitoring systems to a wireless server outside University’s network but occasional network problems continued throughout the monitoring period. The unreliable network connection caused frequent disruptions in data collection and strong bias in the individual electricity consumption data. In addition, network problems caused additional workload to University’s IT specialists whose help was needed whenever the Internet connection went down.  3.1.2 Occupant behavior The inventory of electronic appliances in the four faculty offices revealed a high variability in the number of appliances. Where some offices only had a computer, monitor and printer in them, others were equipped with microwaves, fans, radios and other miscellaneous electronic devices. (Table 1) Consequently, the average plug load level also varied highly between faculty offices. The rooms with highest amount of appliances had a high baseline plug load (plug load level when appliances are plugged in but not used) and higher plug load peaks (highest plug load level when offices are occupied and appliances are in use) during office occupancy. However, the frequent network problems precluded accurate estimation and comparison between per office plug loads and electricity consumptions.  The plug load data also revealed a high variability in occupant schedules during the 9-week monitoring period. The office was assumed to be occupied whenever the plug load level rose above the observed baseline consumption. The monitoring data showed that none of the four occupants followed a traditional office occupancy schedule. The offices were rarely occupied at the same time or for an equal amount of time per day (Figure 3a-d). Only one of the four occupants used the option to control appliance status through the touchscreen monitor and shut off smart strips and sockets when leaving the office. (Figure 3c) None of the participants set schedules for their plug load system to shut off automatically at a certain time of the day. Overall, the consumption patterns of the occupants stayed unchanged during the monitoring period: the occupants who did not use the system features, such as “away” and “home” modes, in the beginning did not develop interest in using them later in the monitoring period either. Respectively, the occupant who used the modes to control office plug load did so throughout the monitoring period. 3.2 Administrator Survey The interviewed University administrators had different levels of involvement in the plug load monitoring study implemented in UW residence halls in 2013. While first interviewee was the overall project manager and the second interviewee responsible for student involvement and recruitment throughout the project, the third one was only responsible for planning the distribution of the plug load monitoring systems. However, when asked about issues related to the design of plug load monitoring studies, all three interviewees mentioned the lack of long-term engagement as the main challenge. According to the interviewees, participants are usually engaged and motivated to reduce their electricity consumption in 018-5 the beginning of any energy intervention, but the involvement fades as the “individuals lose interest” and “everyday life gets in the way”. From the future plug load monitoring studies, the interviewees hoped for more data on occupant behavior and long-term impact measurement both in individual and community level. Two of the three interviewees saw more potential in plug load monitoring in office environments than in residential buildings. One of the respondents supposed that the routines and schedules of an office environment might facilitate the implementation of a plug load monitoring campaign and lead to more long-term occupant engagement. Another administrator speculated that the engagement level in an office environment would be higher as occupants are more exposed to a positive peer pressure than in a residential setting. Overall, respondents considered plug load monitoring as a key method for changing occupants’ consumption behavior, “raising awareness and help drive decision making processes”. However, the respondents did not see plug load monitoring as an efficient way to reduce University’s total electricity consumption as plug load was considered to be “fairly inconsequential in the overall electrical demand of the University buildings”. In addition to the problems related to the long-term occupant engagement, respondents mentioned several miscellaneous factors that challenged the implementation of the residential hall plug load study and might thus cause problems in the future studies as well. As possible technical difficulties, respondents brought up problems with Wi-Fi connection and the resulting interruptions in data collection. One of the administrators also anticipated that if a plug load study was implemented in a much larger scale, the operation and maintenance might become an issue due to university staff’s lack of expertise in the utilized plug load monitoring technology. Other non-technical challenges were mentioned as well. One of them was the lack of focus in the study design: according to one of the respondents, the residential hall experiment did not have a strong objective that would have guided the monitoring process from the beginning to the end. The studied demographic group was also described as challenging: the occupants of the monitored residential halls were mainly freshmen who had just moved on campus and were struggling with their new lifestyle in academic environment. Moreover, they did not generally have high interest in issues related to electricity consumption, as electricity costs were included in their rent. Respondents estimated that student involvement might have been stronger, if situation would have been different.  Table 1: Appliances listing by office Appliance Office 1 Office 2 Office 3 Office 4 Desktop computer ● ● ● ● Monitor 1 ● ● ● ● Monitor 2  ● ○  Fan ● ○ ●  Phone charger ● ○ ○  Phone ○ ● ○  Lamp ●    Printer ● ○ ○ ● Refrigerator ●    Microwave ○  ○  Radio ●    Touchscreen monitor ○ ○ ● ○ Bass  ●                                              *  ○ In the office  ● Monitored  018-6 4 DISCUSSION AND CONCLUSION The feasibility of an office plug load monitoring study was evaluated by implementing a 9-week monitoring campaign in four faculty offices at University of Washington campus. The findings of the 9-week mini study were compared with the results of a residence hall plug load study that was conducted at the same campus in 2013. In addition to monitoring plug loads, the researchers collected data by surveying university administrators that were involved in the prior residence hall plug load study. The administrators were asked about the findings, shortcomings, and recommendations for future studies. The study was able to identify possible challenges and barriers the stakeholders face when deploying plug load monitoring campaigns on campus settings. Moreover, it was able to characterize the load profiles of the faculty/staff offices, even though it failed to make accurate quantitative analyses of the participants’ individual electricity usage. Throughout the 9-week study period, the unreliable network connection caused frequent disruptions in data collection and strong bias in the individual electricity consumption data. The disruptions eventually precluded accurate estimation and comparison between per office plug loads and electricity consumptions. In addition, network problems caused additional workload to University’s IT specialists whose help was needed whenever the Internet connection went down. These findings are in accordance with those of the 2013 residential plug load study, where problems with wireless connections between devices precluded plug load analysis on individual level (Black et al. 2014). Problems with wireless networks have been mentioned by other studies as well: Ghatikar et al. (2013) observed that in addition to being limited by their range, wireless plug load monitoring systems can be prone to high attenuation due to common obstructions in the office environments, such as cubicle separations and concrete walls. As many monitoring systems rely heavily on customers’ wireless networks, such connection problems introduce numerous walk away opportunities and limit wide spread occupant participation (Gilbert et al. 2011). Improvement to the current situation could be received by using more efficient network protocols, i.e. preferring 6lowpan protocol over more limited Zigbee protocol that is currently being used by most wireless monitoring systems (Ellaboudy 2012). The inventory of electronic appliances in the monitored offices revealed a high variability in the number of devices. Consequently, the base consumption and peak plug loads also varied highly between faculty offices. (Figure 3a-3d) Murtagh et al. observed similar variability in their study with weekly energy use of the monitored workstations ranging from near 0 kWh to 21.4 kWh. Both results indicate that even though the work setting for academic office-based researchers is similar to other office settings, there are fundamental differences in occupant behavior and energy use. For instance, depending on their field of study, researchers may have very different needs for IT and other electronic appliances. Moreover, as was observed during this study, some researchers are physically present at their workstations for most of the time, while others are working remotely or sharing their time between several workstations and offices. (Figure 3a-3d) The observed variability in occupant schedules and hours of attendance differs significantly from the widely used ASHRAE 90.1 occupancy profile and implicates that fixed occupancy profiles are not ideal for modeling electricity use and plug loads in academic offices. (ASHRAE 2004) These findings are in line with other recent studies that have suggested updates to the current ASHRAE recommended practice (Bouffaron 2014, Davis & Nutter 2010). The lack of occupant engagement was found to be the main challenge in the implementation of plug load monitoring campaigns. Although all participants of the 9-week office study were taught how to control their electricity use through monitoring equipment, only one of the four occupants used the option. The lack of occupant engagement, especially in longer term, was also mentioned as the main challenge by all of the interviewed University administrators. These results are in great agreement with prior findings of several short- and long-term plug load studies (Hargreaves et al. 2010, Ueno et al. 2006, Murtagh et al. 2013, Ecova 2011) and indicate that the design focus of future plug load monitoring campaigns should be on long-term occupant activation and engagement.  018-8 Acknowledgements The research for this paper was financially supported by the Valle program at University of Washington. We would like to thank the faculties for allowing us to monitor their plug-load use and the university administrators who supportively participated in the interview. References Abrahamse, W., and Steg, L. (2011). “Factors Related to Household Energy Use and Intention to Reduce It: The Role of Psychological and Socio-demographic Variables.” Human Ecology Review, 18(1), p. 30–40. Acker, B., Duarte, C., and Van Den Wymelenberg, K. (2012). Office Space Plug Load Profiles and Energy Saving Interventions. 2012 ACEEE Summer Study on Energy Efficiency in Buildings. ASHRAE (2004). ASHRAE AS. Standard 90.1-2004, energy standard for buildings except low rise residential buildings. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc; 2004. Azar, E., and Menassa, C. (2010). “A Conceptual Framework to Energy Estimation in Buildings Using Agent-Based Modeling.” Proceedings of the 2010 Winter Simulation Conference, IEEE. Black, E., Ha, A., Hall, K., Harms, J., Holen, J., 2014. University of Washington residence hall energy conservation study. Program on the Environment, University of Washington. Bouffaron, P. (2014). Revealing Occupancy Diversity Factors in Buildings Using Sensor Data. Behavior, Energy and Climate Change (BECC) Conference, 2014 conference proceedings, Washington DC. Davis, J.A. III & Nutter, D.W. (2010). Occupancy diversity factors for common university building types. Energy and Buildings, 42 (9), p. 1543-1551. Ecova (2011). Commercial Office Plug Load Savings and Assessment: Executive Summary. Prepared for the California Energy Commission. Ellaboudy, A. (2012). Outlet Power Monitoring Using Wireless Sensor Networks. Technical Report No. UCB/EECS-2012-152. Electrical Engineering and Computer Sciences, University of California at Berkeley. EnergyHub, Inc. (2011). HomeBase – User guide. DOC-DB2-UG-110318. Faruqui, A., Sergici, S., and Sharif, A. (2010). “The Impact of Informational Feedback on Energy Consumption—A Survey of the Experimental Evidence.” Energy, 35, p. 1598–1608. Fischer, C. (2008). “Feedback on Household Electricity Consumption: A Tool for Saving Energy?” Energy Efficiency, 1, p. 79–104. Froehlich, J., Findlater, L., and Landay, J. (2010). “The Design of Eco-Feedback Technology.” CHI 2010: Home Eco Behavior, April 10–15, 2010, Atlanta, GA. Ghatikar, G., Cheung, I., Lanzisera, S. (2013). Miscellaneous and Electronic Loads Energy Efficiency Opportunities for Commercial Buildings: A Collaborative Study by the United States and India. Environmental Energy Technologies Division, Lawrence Berkeley National Laboratory. Gilbert, E., Ekrem, G., Maslowski, R., Schare, S. (2011). Interoperability Lessons from Ongoing Residential Smart Grid Deployments. Grid-Interop Forum 2011.  Hargreaves, T., Nye, M., and Burgess, J. (2010). “Making Energy Visible: A Qualitative Field Study of How Householders Interact with Feedback from Smart Energy Monitors.” Energy Policy, 38, p. 6111–6119. Hargreaves, T., Nye, M., and Burgess, J. (2013). “Keeping Energy Visible? Exploring How Householders Interact with Feedback from Smart Energy Monitors in the Longer Term.” Energy Policy, 52, p. 126–134. Jain, R., Taylor, J. E., and Peschiera, G. (2012). “Assessing Eco-feedback Interface Usage and Design to Drive Energy Efficiency in Buildings.” Energy and Buildings, 48, p. 8–17. Jain, R., Taylor, J. E., and Culligan, P. J. (2013a). “Investigating the Impact Eco-feedback Information Representation Has on Building Occupant Energy Consumption Behavior and Savings.” Energy and Buildings, 64, p. 408–414. Jain, R., Gulbinas, R., Taylor, J. E., and Culligan, P. J. (2013b). “Can Social Influence Drive Energy Savings? Detecting the Impact of Social Influence on the Energy Consumption Behavior of Networked Users Exposed to Normative Eco-feedback.” Energy and Buildings, 66, p. 119–127. 018-9 Jeong, S. H., Gulbinas, R., Jain, R., and Taylor, J. E. (2014). “The Impact of Combined Water and Energy Consumption Eco-feedback on Conservation.” Energy and Buildings, 80, p. 114–119. Kamilaris, A., Kalluri, B., Kondepudi, S., and Wai, T. K. (2014). “A Literature Survey on Measuring Energy Usage for Miscellaneous Electric Loads in Offices and Commercial Buildings.” Renewable and Sustainable Energy Reviews, 34, p. 536–550. Masoso, O. T., and Grobler, L. J. (2010). “The Dark Side of Occupants? Behaviour on Building Energy Use.” Energy and Buildings, 42, p. 173–177.  Murtagh, N., Nati, M., Headley, W. R., Gatersleben, B., Gluhak, A., Imran, M. A., and Uzzell, D. (2013). “Individual Energy Use and Feedback in an Office Setting: A Field Trial.” Energy Policy, 62, p. 717–728. Pereira, L., Quintal, F., Barreto, M., and Nunes, N. J. (2013). “Understanding the Limitations of Eco-feedback: A One-Year Long-Term Study.” A. Holzinger and G. Pasi (Eds.): HCI-KDD 2013, LNCS 7947, p. 237–255. Schober, M. F., Conrad, F. G. (1997). Does Conversational Interviewing Reduce Survey Measurement Error? The Public Opinion Quarterly, 61 (4), p. 576-602. Schweiker, M., and Shukuya, M. (2010). “Comparative Effects of Building Envelope Improvements and Occupant Behavioural Changes on the Energy Consumption for Heating and Cooling.” Energy Policy, 38(6), p. 2976–2986. Siero, F. W., Bakker, A. B., Dekker, G. D., and Van Der Burg, M. T. C (1996). “Changing Organizational Energy Consumption Behavior through Comparative Feedback.” Journal of Environmental Psychology, 16, p. 235–246. Starik, M., and Marcus, A. A. (2000). “Introduction to the Special Research Forum on the Management of Organizations in the Natural Environment: A Field Emerging from Multiple Paths, with Many Challenges Ahead.” The Academy of Management Journal, 43(4), pp. 539–546. Tourangeau, R., Yan, T. (2007). Sensitive questions in surveys. Psychological Bulletin, 133 (5), p. 859-883. Ueno, T., Sano, F., Saeki, O., and Tsuji, K. (2006). “Effectiveness of an Energy-Consumption Information System on Energy Savings in Residential Houses Based on Monitored Data.” Applied Energy, 83, p. 166–183. University of Washington (2012). The Source of Our Power: Electricity at the University of Washington. Available at https://f2.washington.edu/cpo/sites/default/files/file/sustain/uw-energy-power-sources.pdf. U.S. Department of Energy (2014). Annual Energy Reviews 2011–2014. Available at http://www.eia.gov/totalenergy/data/annual/ Vassileva, I., Odlare, M., Wallin, F., and Dahlquist, E. (2012). “The Impact of Consumers’ Feedback Preferences on Domestic Electricity Consumption.” Applied Energy, 93, p. 575–582. Vine, D., Buys, L., and Morris, P. (2013). “The Effectiveness of Energy Feedback for Conservation and Peak Demand: A Literature Review.” Open Journal of Energy Efficiency, 2, p. 7–15.   018-10  5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   THE FEASIBILITY OF PLUG-LOAD MONITORING AND ENERGY-SAVING INTERVENTIONS IN RESIDENTIAL AND OFFICE BUILDINGS ON THE UNIVERSITY OF WASHINGTON CAMPUS  Heta K. Kosonen1,3, Amy A. Kim2  1 Department of Civil and Environmental Engineering, University of Washington, USA 2 Department of Civil and Environmental Engineering, University of Washington, USA 3 hetak@uw.edu  Abstract: The University of Washington (UW) is aiming to reduce the overall electricity consumption on campus as part of its Climate Action Plan launched in 2009. To achieve this goal, UW installed 216 smart grid meters and automatic heating, ventilation, and cooling control systems across the entire campus and acquired over 200 sets of plug-load monitoring equipment. The university used the smart grid data and the monitored plug-load data to test how occupants in selected residence halls responded to receiving detailed information about their energy usage patterns, its environmental impacts, and associated costs. The experiment demonstrated that in residence halls, plug-load monitoring did not have any significant impact on the occupants’ electricity consumption. Hence, there is still a need to further assess which strategies are effective in achieving long-term electricity reduction goals for the university. The goal of this study was to conduct a comparative analysis by replicating the plug-load analysis conducted in residence halls in a faculty/staff office setting. The study entailed interviewing university administrators that were involved in the residence hall plug-load study. Interviewees were asked questions about the findings, shortcomings, and recommendations for future studies. Also, this study characterized the load profiles of the faculty/staff offices by monitoring the plug-load consumption in four offices for nine weeks and explored plug-load reduction interventions applicable to office settings. The study found that the unreliable network connection caused frequent disruptions in data collection and strong bias in the individual electricity consumption data. The inventory of electronic appliances in the monitored offices revealed a high variability in the number of devices which lead to variations in base consumption and peak plug loads between faculty offices, and lack of occupant engagement was found to be the main challenge in the implementation of plug load monitoring campaigns. The results provide universities around the country with valuable information and insights on how to design and implement an on campus plug-load reduction intervention with quantifiable energy-saving potential. 1 BACKGROUND Worldwide reductions in greenhouse gas emissions are needed to mitigate the most severe impacts of climate change. Commercial buildings account for roughly 18 percent of the total annual energy use in the United States (US DOE 2014). Therefore, sustainability initiatives at workplaces offer a substantial opportunity to reduce the harmful effects on the environment. The University of Washington (UW) is aiming to reduce the overall electricity consumption on campus as part of its Climate Action Plan launched in 2009. Currently, miscellaneous electricity loads (e.g., computers and office equipment) from offices account for approximately 12 percent of the total electricity 18-1 used in the UW, which means that an average of 36 million kilowatt hours of electricity is used annually on plug loads (University of Washington 2012). According to the results presented in recent literature, a well-designed energy intervention in the UW offices has the potential to save up to $360,000 of UW’s annual electricity costs. Furthermore, reducing plug loads would not only help the university to reach its carbon neutrality goal but would also improve the power quality on the local grid and moderate peak electrical demand in the Seattle area. Traditionally, energy-conservation efforts in office environments have been implemented through technological or operational modifications (Starik & Marcus 2000). However, humans are the main operators of technology, and a failure of the human component can fail the entire energy-efficiency initiative. Thus, instead of focusing solely on technological solutions, recent research has shifted to investigate the effects of occupant behavior on building energy use (Fischer 2008, Azar & Menassa 2010, Masoso & Grobler 2010, Schweiker & Shukuya 2010, Kamilaris et al. 2014). The results of these studies have shown that the energy-saving potential of behavioral change is comparable to, and even higher than, that of technological solutions (Masoso &Grobler 2010, Schweiker & Shukuya 2010). Some estimates even suggest that the occupants control or impact up to 50 percent of a building’s energy use and that changing occupant behavior patterns gives the most effective reductions in energy use (Kamilaris et al. 2014). Changing occupant behavior in offices and other commercial buildings is not without its challenges. A wide variety of studies have looked into the different types of interventions that could most effectively result in electricity savings. One of the most applied measures of impacting and controlling occupant energy use is giving occupants regular feedback on their energy usage patterns (Jain et al. 2012, Jeong et al. 2014, Gulbinas et al. 2014, Hargreaves et al. 2010, Hargreaves et al. 2013, Pereira et al. 2013, Froehlich et al. 2010, Vine et al. 2013). Because feedback frequency and accessibility have been found to correlate positively with the impact on energy reductions (Abrahamse & Steg 2011), most of the recent research on occupant behavior has used real-time monitoring solutions (Jain et al. 2012, Gulbinas et al. 2014, Jain et al. 2013a, Jain et al. 2013b, Ueno et al. 2006). Various studies have shown that frequent feedback is generally effective and correlates negatively with the energy consumption rate (Faruqui et al. 2010, Siero et al. 1996, Vassileva et al. 2012, Murtagh et al. 2013). However, its effects are often temporary, as the engagement of the participants has been repeatedly observed to reduce over time (Hargreaves et al. 2010, Ueno et al. 2006, Murtagh et al. 2013). Furthermore, not everyone is interested in receiving feedback on their electricity consumption: In their study, Murtagh et al. found out that 41% of the participants did not access their individualized feedback even once. These results indicate that in order to design effective electricity interventions with consumption feedback, the focus should be in long-term participant engagement. When implemented successfully, high-frequency electricity feedback can result in total electricity savings of about 20 percent (Murtagh et al. 2013, Acker et al. 2012, Ecova 2011). UW has already installed 216 smart grid meters and automatic heating, ventilation, and cooling control systems across the entire campus and acquired over 200 sets of plug-load monitoring equipment. In 2013, the university used the smart grid data and the monitored plug-load data to test how occupants in selected residence halls responded to receiving detailed information about their energy usage patterns, its environmental impacts, and associated costs. The project team examined electricity use in each of the buildings over a ten-week period in order to understand which intervention, a technology intervention or education intervention, would have a greater effect (if any) on floor-wide energy consumption. The experiment demonstrated that in residence halls, neither educational nor technical plug-load reduction interventions had any significant impact on the occupants’ electricity consumption: Occupants appeared to have higher energy use throughout the study, and educational intervention failed to produce statistically significant results. The research group listed small sample size, inexpensive energy, subjects who do not pay individual energy bills and technical difficulties as some of the factors that may have contributed to these results. Despite the inconclusive results, the team suggests that University has potential to educate students and successfully reduce energy use though other approaches. However, the use of plug load monitoring systems was not recommended as it was found to be an expensive and ineffective tool for changing energy behavior in the context of the University’s residence halls (Black et al. 2014). 018-2 The goal of our study is to conduct a comparative analysis by replicating the plug load monitoring campaign conducted in residence halls in a faculty/staff office setting. In addition to collecting monitoring data, we conducted a survey of the university administrators that were involved in the 2013 residence hall plug-load study to learn more about the challenges and opportunities related to plug load monitoring on campus setting. The study will serve as a preliminary study for a plug load monitoring campaign that will be implemented in one of the office buildings at UW campus later in 2015. We believe that the results will provide universities around the country with valuable information on how to design and implement an on-campus plug-load reduction intervention with quantifiable energy-saving potential. 2 METHODS 2.1 Plug load monitoring  2.1.1 Equipment Plug load monitoring systems with control capability were installed in four faculty offices. The systems consisted of smart power sockets and strips, a Wi-Fi-connected touchscreen monitor with control capability over smart sockets and strips, and an online user account for data collection. (Figure 1) A total number of 20 appliances were plugged into the smart sockets and strips that were connected to the touchscreen monitors over Wi-Fi. High power appliances, such as refrigerators, fans and microwaves had to be excluded from the study as the monitoring system only supports devices with up to a maximum of 15 amps (EnergyHub Inc. 2011). In addition, University’s IT staff requested that desktop computers were kept on at all times to allow for software and security updates.  Figure 1. Set-up of the plug load monitoring system 2.1.2 Installation and education The monitoring systems were set up over a period of one week. Prior to installation, the office occupants conducted an inventory of their electronic appliances together with the research staff and identified the appliances that were to be connected to the monitoring system. In addition, each appliance was given a status on the basis of occupant’s requests: If appliance was given an “always on” status, it would stay on even if the smart strips and sockets were turned off. Other appliances with an “on-off” status would turn off normally when the power to the strips and sockets was cut off.  The installation process consisted of three phases. In first phase, sockets and strips were connected to the touchscreen monitor by using strip- and socket-specific set-up codes. In the second phase, the electronic appliances in each socket and strip were named in order to allow appliance-by-appliance electricity monitoring. In the last phase, the status of each appliance was determined by typing the information in the touchscreen monitor. In order to minimize the disturbance to the occupants, installation work was completed when offices were unoccupied.  018-3 3 RESULTS 3.1 Plug load monitoring 3.1.1 Monitoring system installation & operation Depending on the amount of electronic devices in the office, the installation process took approximately 30 to 45 minutes per office. After installation, the exact locations and identifying information of all smart sockets, strips and touchscreen monitors were collected to a directory, which was kept up to date about the performance of the study equipment through the monitoring period. Apart from couple of malfunctioning smart sockets, the monitoring hardware functioned as expected and did not require maintenance over the 9-week study period. The total time used for hardware installation and operation was approximately 1-1.5 hours per office per 9 weeks. The initial plan was to connect all monitoring systems to the University’s Wi-Fi but after unsuccessful attempts in all office rooms it became evident that the touchscreen monitors did not communicate with the server: users were unable to access their devices remotely and no electricity consumption data were saved to the system database online. The issue was temporarily resolved by connecting the monitoring systems to a wireless server outside University’s network but occasional network problems continued throughout the monitoring period. The unreliable network connection caused frequent disruptions in data collection and strong bias in the individual electricity consumption data. In addition, network problems caused additional workload to University’s IT specialists whose help was needed whenever the Internet connection went down.  3.1.2 Occupant behavior The inventory of electronic appliances in the four faculty offices revealed a high variability in the number of appliances. Where some offices only had a computer, monitor and printer in them, others were equipped with microwaves, fans, radios and other miscellaneous electronic devices. (Table 1) Consequently, the average plug load level also varied highly between faculty offices. The rooms with highest amount of appliances had a high baseline plug load (plug load level when appliances are plugged in but not used) and higher plug load peaks (highest plug load level when offices are occupied and appliances are in use) during office occupancy. However, the frequent network problems precluded accurate estimation and comparison between per office plug loads and electricity consumptions.  The plug load data also revealed a high variability in occupant schedules during the 9-week monitoring period. The office was assumed to be occupied whenever the plug load level rose above the observed baseline consumption. The monitoring data showed that none of the four occupants followed a traditional office occupancy schedule. The offices were rarely occupied at the same time or for an equal amount of time per day (Figure 3a-d). Only one of the four occupants used the option to control appliance status through the touchscreen monitor and shut off smart strips and sockets when leaving the office. (Figure 3c) None of the participants set schedules for their plug load system to shut off automatically at a certain time of the day. Overall, the consumption patterns of the occupants stayed unchanged during the monitoring period: the occupants who did not use the system features, such as “away” and “home” modes, in the beginning did not develop interest in using them later in the monitoring period either. Respectively, the occupant who used the modes to control office plug load did so throughout the monitoring period. 3.2 Administrator Survey The interviewed University administrators had different levels of involvement in the plug load monitoring study implemented in UW residence halls in 2013. While first interviewee was the overall project manager and the second interviewee responsible for student involvement and recruitment throughout the project, the third one was only responsible for planning the distribution of the plug load monitoring systems. However, when asked about issues related to the design of plug load monitoring studies, all three interviewees mentioned the lack of long-term engagement as the main challenge. According to the interviewees, participants are usually engaged and motivated to reduce their electricity consumption in 018-5 the beginning of any energy intervention, but the involvement fades as the “individuals lose interest” and “everyday life gets in the way”. From the future plug load monitoring studies, the interviewees hoped for more data on occupant behavior and long-term impact measurement both in individual and community level. Two of the three interviewees saw more potential in plug load monitoring in office environments than in residential buildings. One of the respondents supposed that the routines and schedules of an office environment might facilitate the implementation of a plug load monitoring campaign and lead to more long-term occupant engagement. Another administrator speculated that the engagement level in an office environment would be higher as occupants are more exposed to a positive peer pressure than in a residential setting. Overall, respondents considered plug load monitoring as a key method for changing occupants’ consumption behavior, “raising awareness and help drive decision making processes”. However, the respondents did not see plug load monitoring as an efficient way to reduce University’s total electricity consumption as plug load was considered to be “fairly inconsequential in the overall electrical demand of the University buildings”. In addition to the problems related to the long-term occupant engagement, respondents mentioned several miscellaneous factors that challenged the implementation of the residential hall plug load study and might thus cause problems in the future studies as well. As possible technical difficulties, respondents brought up problems with Wi-Fi connection and the resulting interruptions in data collection. One of the administrators also anticipated that if a plug load study was implemented in a much larger scale, the operation and maintenance might become an issue due to university staff’s lack of expertise in the utilized plug load monitoring technology. Other non-technical challenges were mentioned as well. One of them was the lack of focus in the study design: according to one of the respondents, the residential hall experiment did not have a strong objective that would have guided the monitoring process from the beginning to the end. The studied demographic group was also described as challenging: the occupants of the monitored residential halls were mainly freshmen who had just moved on campus and were struggling with their new lifestyle in academic environment. Moreover, they did not generally have high interest in issues related to electricity consumption, as electricity costs were included in their rent. Respondents estimated that student involvement might have been stronger, if situation would have been different.  Table 1: Appliances listing by office Appliance Office 1 Office 2 Office 3 Office 4 Desktop computer ● ● ● ● Monitor 1 ● ● ● ● Monitor 2  ● ○  Fan ● ○ ●  Phone charger ● ○ ○  Phone ○ ● ○  Lamp ●    Printer ● ○ ○ ● Refrigerator ●    Microwave ○  ○  Radio ●    Touchscreen monitor ○ ○ ● ○ Bass  ●                                              *  ○ In the office  ● Monitored  018-6 4 DISCUSSION AND CONCLUSION The feasibility of an office plug load monitoring study was evaluated by implementing a 9-week monitoring campaign in four faculty offices at University of Washington campus. The findings of the 9-week mini study were compared with the results of a residence hall plug load study that was conducted at the same campus in 2013. In addition to monitoring plug loads, the researchers collected data by surveying university administrators that were involved in the prior residence hall plug load study. The administrators were asked about the findings, shortcomings, and recommendations for future studies. The study was able to identify possible challenges and barriers the stakeholders face when deploying plug load monitoring campaigns on campus settings. Moreover, it was able to characterize the load profiles of the faculty/staff offices, even though it failed to make accurate quantitative analyses of the participants’ individual electricity usage. Throughout the 9-week study period, the unreliable network connection caused frequent disruptions in data collection and strong bias in the individual electricity consumption data. The disruptions eventually precluded accurate estimation and comparison between per office plug loads and electricity consumptions. In addition, network problems caused additional workload to University’s IT specialists whose help was needed whenever the Internet connection went down. These findings are in accordance with those of the 2013 residential plug load study, where problems with wireless connections between devices precluded plug load analysis on individual level (Black et al. 2014). Problems with wireless networks have been mentioned by other studies as well: Ghatikar et al. (2013) observed that in addition to being limited by their range, wireless plug load monitoring systems can be prone to high attenuation due to common obstructions in the office environments, such as cubicle separations and concrete walls. As many monitoring systems rely heavily on customers’ wireless networks, such connection problems introduce numerous walk away opportunities and limit wide spread occupant participation (Gilbert et al. 2011). Improvement to the current situation could be received by using more efficient network protocols, i.e. preferring 6lowpan protocol over more limited Zigbee protocol that is currently being used by most wireless monitoring systems (Ellaboudy 2012). The inventory of electronic appliances in the monitored offices revealed a high variability in the number of devices. Consequently, the base consumption and peak plug loads also varied highly between faculty offices. (Figure 3a-3d) Murtagh et al. observed similar variability in their study with weekly energy use of the monitored workstations ranging from near 0 kWh to 21.4 kWh. Both results indicate that even though the work setting for academic office-based researchers is similar to other office settings, there are fundamental differences in occupant behavior and energy use. For instance, depending on their field of study, researchers may have very different needs for IT and other electronic appliances. Moreover, as was observed during this study, some researchers are physically present at their workstations for most of the time, while others are working remotely or sharing their time between several workstations and offices. (Figure 3a-3d) The observed variability in occupant schedules and hours of attendance differs significantly from the widely used ASHRAE 90.1 occupancy profile and implicates that fixed occupancy profiles are not ideal for modeling electricity use and plug loads in academic offices. (ASHRAE 2004) These findings are in line with other recent studies that have suggested updates to the current ASHRAE recommended practice (Bouffaron 2014, Davis & Nutter 2010). The lack of occupant engagement was found to be the main challenge in the implementation of plug load monitoring campaigns. Although all participants of the 9-week office study were taught how to control their electricity use through monitoring equipment, only one of the four occupants used the option. The lack of occupant engagement, especially in longer term, was also mentioned as the main challenge by all of the interviewed University administrators. These results are in great agreement with prior findings of several short- and long-term plug load studies (Hargreaves et al. 2010, Ueno et al. 2006, Murtagh et al. 2013, Ecova 2011) and indicate that the design focus of future plug load monitoring campaigns should be on long-term occupant activation and engagement.  018-8 Acknowledgements The research for this paper was financially supported by the Valle program at University of Washington. We would like to thank the faculties for allowing us to monitor their plug-load use and the university administrators who supportively participated in the interview. References Abrahamse, W., and Steg, L. (2011). “Factors Related to Household Energy Use and Intention to Reduce It: The Role of Psychological and Socio-demographic Variables.” Human Ecology Review, 18(1), p. 30–40. Acker, B., Duarte, C., and Van Den Wymelenberg, K. (2012). Office Space Plug Load Profiles and Energy Saving Interventions. 2012 ACEEE Summer Study on Energy Efficiency in Buildings. ASHRAE (2004). ASHRAE AS. Standard 90.1-2004, energy standard for buildings except low rise residential buildings. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc; 2004. Azar, E., and Menassa, C. (2010). “A Conceptual Framework to Energy Estimation in Buildings Using Agent-Based Modeling.” Proceedings of the 2010 Winter Simulation Conference, IEEE. Black, E., Ha, A., Hall, K., Harms, J., Holen, J., 2014. University of Washington residence hall energy conservation study. Program on the Environment, University of Washington. Bouffaron, P. (2014). Revealing Occupancy Diversity Factors in Buildings Using Sensor Data. Behavior, Energy and Climate Change (BECC) Conference, 2014 conference proceedings, Washington DC. Davis, J.A. III & Nutter, D.W. (2010). Occupancy diversity factors for common university building types. Energy and Buildings, 42 (9), p. 1543-1551. Ecova (2011). Commercial Office Plug Load Savings and Assessment: Executive Summary. Prepared for the California Energy Commission. Ellaboudy, A. (2012). Outlet Power Monitoring Using Wireless Sensor Networks. Technical Report No. UCB/EECS-2012-152. Electrical Engineering and Computer Sciences, University of California at Berkeley. EnergyHub, Inc. (2011). HomeBase – User guide. DOC-DB2-UG-110318. Faruqui, A., Sergici, S., and Sharif, A. (2010). “The Impact of Informational Feedback on Energy Consumption—A Survey of the Experimental Evidence.” Energy, 35, p. 1598–1608. Fischer, C. (2008). “Feedback on Household Electricity Consumption: A Tool for Saving Energy?” Energy Efficiency, 1, p. 79–104. Froehlich, J., Findlater, L., and Landay, J. (2010). “The Design of Eco-Feedback Technology.” CHI 2010: Home Eco Behavior, April 10–15, 2010, Atlanta, GA. Ghatikar, G., Cheung, I., Lanzisera, S. (2013). Miscellaneous and Electronic Loads Energy Efficiency Opportunities for Commercial Buildings: A Collaborative Study by the United States and India. Environmental Energy Technologies Division, Lawrence Berkeley National Laboratory. Gilbert, E., Ekrem, G., Maslowski, R., Schare, S. (2011). Interoperability Lessons from Ongoing Residential Smart Grid Deployments. Grid-Interop Forum 2011.  Hargreaves, T., Nye, M., and Burgess, J. (2010). “Making Energy Visible: A Qualitative Field Study of How Householders Interact with Feedback from Smart Energy Monitors.” Energy Policy, 38, p. 6111–6119. Hargreaves, T., Nye, M., and Burgess, J. (2013). “Keeping Energy Visible? Exploring How Householders Interact with Feedback from Smart Energy Monitors in the Longer Term.” Energy Policy, 52, p. 126–134. Jain, R., Taylor, J. E., and Peschiera, G. (2012). “Assessing Eco-feedback Interface Usage and Design to Drive Energy Efficiency in Buildings.” Energy and Buildings, 48, p. 8–17. Jain, R., Taylor, J. E., and Culligan, P. J. (2013a). “Investigating the Impact Eco-feedback Information Representation Has on Building Occupant Energy Consumption Behavior and Savings.” Energy and Buildings, 64, p. 408–414. Jain, R., Gulbinas, R., Taylor, J. E., and Culligan, P. J. (2013b). “Can Social Influence Drive Energy Savings? Detecting the Impact of Social Influence on the Energy Consumption Behavior of Networked Users Exposed to Normative Eco-feedback.” Energy and Buildings, 66, p. 119–127. 018-9 Jeong, S. H., Gulbinas, R., Jain, R., and Taylor, J. E. (2014). “The Impact of Combined Water and Energy Consumption Eco-feedback on Conservation.” Energy and Buildings, 80, p. 114–119. Kamilaris, A., Kalluri, B., Kondepudi, S., and Wai, T. K. (2014). “A Literature Survey on Measuring Energy Usage for Miscellaneous Electric Loads in Offices and Commercial Buildings.” Renewable and Sustainable Energy Reviews, 34, p. 536–550. Masoso, O. T., and Grobler, L. J. (2010). “The Dark Side of Occupants? Behaviour on Building Energy Use.” Energy and Buildings, 42, p. 173–177.  Murtagh, N., Nati, M., Headley, W. R., Gatersleben, B., Gluhak, A., Imran, M. A., and Uzzell, D. (2013). “Individual Energy Use and Feedback in an Office Setting: A Field Trial.” Energy Policy, 62, p. 717–728. Pereira, L., Quintal, F., Barreto, M., and Nunes, N. J. (2013). “Understanding the Limitations of Eco-feedback: A One-Year Long-Term Study.” A. Holzinger and G. Pasi (Eds.): HCI-KDD 2013, LNCS 7947, p. 237–255. Schober, M. F., Conrad, F. G. (1997). Does Conversational Interviewing Reduce Survey Measurement Error? The Public Opinion Quarterly, 61 (4), p. 576-602. Schweiker, M., and Shukuya, M. (2010). “Comparative Effects of Building Envelope Improvements and Occupant Behavioural Changes on the Energy Consumption for Heating and Cooling.” Energy Policy, 38(6), p. 2976–2986. Siero, F. W., Bakker, A. B., Dekker, G. D., and Van Der Burg, M. T. C (1996). “Changing Organizational Energy Consumption Behavior through Comparative Feedback.” Journal of Environmental Psychology, 16, p. 235–246. Starik, M., and Marcus, A. A. (2000). “Introduction to the Special Research Forum on the Management of Organizations in the Natural Environment: A Field Emerging from Multiple Paths, with Many Challenges Ahead.” The Academy of Management Journal, 43(4), pp. 539–546. Tourangeau, R., Yan, T. (2007). Sensitive questions in surveys. Psychological Bulletin, 133 (5), p. 859-883. Ueno, T., Sano, F., Saeki, O., and Tsuji, K. (2006). “Effectiveness of an Energy-Consumption Information System on Energy Savings in Residential Houses Based on Monitored Data.” Applied Energy, 83, p. 166–183. University of Washington (2012). The Source of Our Power: Electricity at the University of Washington. Available at https://f2.washington.edu/cpo/sites/default/files/file/sustain/uw-energy-power-sources.pdf. U.S. Department of Energy (2014). Annual Energy Reviews 2011–2014. Available at http://www.eia.gov/totalenergy/data/annual/ Vassileva, I., Odlare, M., Wallin, F., and Dahlquist, E. (2012). “The Impact of Consumers’ Feedback Preferences on Domestic Electricity Consumption.” Applied Energy, 93, p. 575–582. Vine, D., Buys, L., and Morris, P. (2013). “The Effectiveness of Energy Feedback for Conservation and Peak Demand: A Literature Review.” Open Journal of Energy Efficiency, 2, p. 7–15.   018-10  THE FEASIBILITY OF PLUG-LOAD MONITORING AND ENERGY-SAVING INTERVENTIONS IN RESIDENTIAL AND OFFICE BUILDINGS ON THE UNIVERSITY OF WASHINGTON CAMPUSHeta K. Kosonen (MSc), Amy. A. Kim (PhD)WE NEED TO CHANGE THE WAY WE CONSUME ENERGY.Source: IPCC, 2014. 5th assessment report.March 2015 (NOAA)SUSTAINABILITY INITIATIVES AT WORKPLACES OFFER A SUBSTANTIAL OPPORTUNITY TO REDUCE GHG EMISSIONS.The building sector is responsible for 39.7% of the total annual energy consumption in the U.S. (EIA 2014).THE SAVINGS POTENTIAL LIES IN PLUG LOADSPlug load Electronic devices not responsible for zone heating and cooling, water heating, or lighting.  Office information technology (IT) equipment Personal appliances (e.g. coffee machines, table fans, and personal space heaters) HVAC41%Plug loads13%Lighting27%Water heating3%Other9%Cooking2%Refrigeration5%WHAT WE KNOW ABOUT PLUG LOAD INTERVENTIONS“Energy-saving potential of behavioral change is comparable to, and even higher than, that of technological solutions” Masoso &Grobler 2010, Schweiker & Shukuya 2010“Occupants control or impact up to 50 percent of a building’s energy use and that changing occupant behavior patterns gives the most effective reductions in energy use” Kamilaris et al. 2014“Occupant energy use can be impacted and controlled by giving occupants regular feedback on their energy usagepatterns.” Jain et al. 2012, Jeong et al. 2014, Gulbinas et al. 2014, Hargreaves et al. 2010, Hargreaves et al. 2013, Pereira et al. 2013, Froehlich et al. 2010, Vine et al. 2013“Frequent feedback is generally effective and correlates negatively with the energy consumption rate.” Faruqui et al. 2010, Siero et al. 1996, Vassileva et al. 2012, Murtagh et al. 2013WHAT WE STILL NEED TO FIGURE OUTWho maintains and operates the monitoring systems?What do we do with the data?How do we keep occupants engaged?What are the ingredients of a successful energy intervention?What happens in the long run?What are the main challenges and barriers related to plug load interventions?PLUG LOAD CONTROL AT THE UNIVERSITY OF WASHINGTONResidential hall studyWinter 2014 10 weeks of monitoring  Technology vs. educational intervention  Results showed no reductions in average energy usePLUG LOAD CONTROL AT THE UNIVERSITY OF WASHINGTON Faculty office studySummer 2014 Identify key issues related to plug load reductions on campus Characterize load profiles Assess the feasibility of plug load interventions in academic offices Pre-study for a larger scale energy interventionMETHODSTouchscreen monitor with Wi-FiSmart stripsSmart socketsOnline control& data collectionMONITORING SYSTEMINSTALLATION AND EDUCATION1. Appliance inventory 2. Setting appliance statuses- Always on, on-off3. System installation- Smart sockets and strips, monitor4. Guidance on system control- Introduction to short-cut commandsADMINISTRATOR SURVEYCONCEPTFeasibility of a plug load monitoring campaign at  UW campus officesDIMENSIONHuman resourcesSUB-DIMENSIONAttitudes towards energy interventionsSUB-DIMENSIONManpowerDIMENSIONStudy designSUB-DIMENSIONTechnical challengesSUB-DIMENSIONMethodological challenges Personal interviews and e-mail questionnaires.  Participants university employees who had been involved in the residence hall plug load study in 2014  A set of 14 questions was created on the basis of the following research topics WHAT WE LEARNED1. ENERGY DATA COLLECTION IS NOT WITHOUT ITS CHALLENGES.> Installation process took approximately 30 to 45 minutes per office. – Little equipment maintenance was required over the 9-week study period. > Occasional network problems throughout the monitoring period.- Frequent disruptions in data collection and strong bias in the individual electricity consumption data. - Additional workload to University’s IT specialists2. ENERGY CONSUMPTION VARIES HIGHLY WITHIN SIMILAR WORK STATIONS.Appliance Office 1 Office 2 Office 3 Office 4Desktop computer ● ● ● ●Monitor 1 ● ● ● ●Monitor 2 ● ○Fan ● ○ ●Phone charger ● ○ ○Phone ○ ● ○Lamp ●Printer ● ○ ○ ●Refrigerator ●Microwave ○ ○Radio ●Touchscreen monitor ○ ○ ● ○Bass ●TOTAL 11 9 9 4○ In the office  ● Monitored3. (ACADEMIC) EMPLOYEES DO NOT FOLLOW TRADITIONAL OCCUPANCY SCHEDULES.4. THE PRESENCE OF PLUG LOAD MONITORING EQUIPMENT ALONE DOES NOT AFFECT CONSUMPTION.- Not everyone is interested- Only one occupant actively used control options- Findings support the results presented in prior studies - Consumption patterns of the occupants stayed unchanged during the monitoring period- Occupants who did not use plug load control in the beginning of the study did not develop interest in using it later in the study- Occupant who used control option kept using it until the end of the study5. LONG-TERM ENGAGEMENT IS A KEY CHALLENGE IN ENERGY BEHAVIOR INTERVENTIONS.- “everyday life gets in the way” - All interviewees mentioned problems with occupant engagement & hoped for more data on long-term impacts of energy interventionsNEXT STEPS- Research on office occupant behavior in high performance buildings- 9-month energy intervention campaign starting on the UW campus- Plug load monitoring- Behavior interventions- Focus on long-term occupant engagement & data management- Effectiveness of different forms of consumption feedback- Simultaneous application of various intervention methodsTHANK YOU.Heta Kosonen Amy A. KimPhD Student Assistant ProfessorCEE department CEE department hetak@uw.edu amyakim@uw.edu

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52660.1-0076491/manifest

Comment

Related Items