Non-Invasive Personalized Normative Messaging Intervention for the Reduction of Household Energy Consumption

1705273(李)。美国家庭化石燃料消耗约占一次能源消耗和二氧化碳当量排放的22%。因此，该项目的一个目标是确定广泛适用的干预方法，这些方法可能能够促进对环境负责的行为。本研究的目的是通过开发和验证基于非侵入性数据挖掘的行为干预框架，通过在密歇根州荷兰的家庭进行现场实验，促进对个性化规范比较组如何影响规范反馈干预有效性的理解。这项研究的成功实施将通过对个性化规范反馈提供详细的初步了解，加强有利于环境的反馈干预设计，从而大大有助于减少建筑环境的有害排放。具体的研究目标是：1)利用非侵入性技术，根据每小时能源使用数据，将家庭划分为几个有意义的群体，分享相似的消费模式；2)利用现成的消费数据，生成个性化的标准能源使用反馈，并对其有效性进行实证评估。这项研究的目标是深入了解个性化规范信息的可靠性，以及能源使用行为如何以及在哪里发生变化。此外，它试图发现个性化规范反馈对家庭能源消费的长期影响，并确定描述性规范参考小组对能源消费规范遵守和能源使用的影响。所有这些都可以通过提出规范的能源反馈理论来为降低住宅能耗做出贡献。关于个性化规范报文在短期和长期内如何以及在哪里改变居住者行为的研究结果，对能源报告、政策制定以及实现州和国家能源削减目标具有重要影响。此外，从能源使用方面的个性化规范信息中吸取的经验教训可以很容易地应用于其他有利于环境的行为(例如，水的消耗)。具体地说，所提出的非侵入式个性化反馈消息收发框架可能能够在高级公用事业网络中广泛部署。

1705273 (Lee). Household fossil fuel consumption in the U.S. is responsible for approximately 22% of primary energy consumption and CO2 equivalent emissions. Consequently, it is an objective of this project to identify widely applicable intervention methods potentially capable of promoting environmentally responsible behaviors. The goal of this research is to advance understanding of how personalized normative comparison groups influence the effectiveness of normative feedback interventions through the development and validation of a non-invasive data mining-based behavior intervention framework, with field experiments conducted in homes in Holland, Michigan. Successful implementation of the research would significantly contribute to reducing harmful emissions from the built environment through the enhancement of pro-environmental feedback intervention design by providing a detailed first look into personalized normative feedback. Specific research objectives are: 1) to classify households into several meaningful groups sharing similar consumption patterns on the basis of hourly energy usage data, using non-invasive techniques; and 2) to generate and then empirically evaluate the effectiveness of personalized normative energy use feedback created with the use of readily available consumption data. The research is targeted to provide an in-depth understanding of both the reliability of personalized normative messages, and how and where energy use behavior changes. Further, it seeks to discover long-term effects of personalized normative feedback on household energy consumption and identify the effect of descriptive norm reference groups on energy consumption norm adherence and energy use. All of these could contribute to residential energy use reduction by advancing the theories for normative energy feedback. The findings on how and where personalized normative messaging changes occupant behavior in both the short and long term have important implications for energy reporting, policy making, and meeting of state and national energy reduction goals. In addition, lessons learned on personalized normative messaging on energy use can be readily applied to other pro-environmental behaviors (e.g., water consumption). Particularly, the proposed non-intrusive personalized feedback messaging framework may be capable of wide-scale deployment in advanced utility networks.