COnsumer-centric Privacy in smart Energy gridS

智能能源网格中以消费者为中心的隐私

• 批准号: EP/N021738/1
• 负责人: Deniz Gunduz
• 金额: $ 43.69万
• 依托单位: Imperial College London
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2015
• 资助国家: 英国
• 起止时间: 2015 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FN021738%2F1
• 关键词: COnsumer; centric; Privacy; smart; Energy

Smart meters have the capability to measure and record consumption data at a high time resolution and communicate such data to the energy provider. This provides the opportunity to better monitor and control the power grid and to enable demand response at the residential level. This not only improves the reliability of grid operations but also constitutes a key enabler to integrate variable renewable generation, such as wind or solar. However, the communication of high resolution consumption data also poses privacy risks as such data allows the utility, or a third party, to derive detailed information about consumer behavior. Hence, the main research objective of COPES is to develop new technologies to protect consumer privacy, while not sacrificing the "smartness", i.e., advanced control and monitoring functionalities. The core idea is to overlay the original consumption pattern with additional physical consumption or generation, thereby hiding the consumer privacy sensitive consumption. The means to achieve this include the usage of storage, small scale distributed generation and/or elastic energy consumptions. Hence, COPES proposes and develops a radically new approach to alter the physical energy flow, instead of purely relying on encryption of meter readings, which provides protection against third party intruders but does not prevent the use of this data by the energy provider. In order to efficiently hide consumption information, intelligent decisions and strategies on when to charge/discharge the storage, which energy source to tap into, need to be made in real time. Therefore, in this project, algorithms based on and extending upon differential privacy, information and detection theoretic first principles that allow efficient use of physical capabilities to alter the overall consumption measured by the smart meters will be developed. Since these resources can also be used to minimize the electricity bill or increase the integration of renewables, trade-offs between these objectives and privacy will be studied and combined into a holistic privacy guaranteeing house energy management system. Implementations on multiple small test systems will serve as a proof of concept of the proposed methods.

智能电表能够以高时间分辨率测量和记录消耗数据，并将这些数据传送给能源供应商。这提供了更好地监测和控制电网并实现住宅级需求响应的机会。这不仅提高了电网运营的可靠性，而且是整合风能或太阳能等可变可再生能源发电的关键推动因素。然而，高分辨率消费数据的通信也带来隐私风险，因为这种数据允许公用事业公司或第三方导出关于消费者行为的详细信息。因此，COPES的主要研究目标是开发新技术来保护消费者隐私，同时不牺牲“智能”，即，先进的控制和监控功能。其核心思想是用额外的物理消费或生成覆盖原始消费模式，从而隐藏消费者隐私敏感的消费。实现这一目标的手段包括使用存储、小规模分布式发电和/或弹性能耗。因此，COPES提出并开发了一种全新的方法来改变物理能量流，而不是纯粹依赖于仪表读数的加密，这提供了对第三方入侵者的保护，但不会阻止能源供应商使用这些数据。为了有效地隐藏消耗信息，需要真实的实时地做出关于何时对存储器充电/放电、利用哪个能量源的智能决策和策略。因此，在本项目中，将开发基于并扩展差分隐私、信息和检测理论第一原理的算法，这些算法允许有效利用物理能力来改变智能电表测量的总体消耗。由于这些资源也可以用来最大限度地减少电费或增加可再生能源的整合，这些目标和隐私之间的权衡将被研究并结合到一个整体的隐私保障家庭能源管理系统。在多个小型测试系统上的实现将作为所提出方法的概念证明。

项目成果

Smart Meter Privacy With Renewable Energy and an Energy Storage Device

• DOI: 10.1109/tifs.2017.2744601
• 发表时间: 2018-01-01
• 期刊: IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
• 影响因子: 6.8
• 作者: Giaconi, Giulio;Gunduz, Deniz;Poor, H. Vincent
• 通讯作者: Poor, H. Vincent

Smart meter privacy with an energy harvesting device and instantaneous power constraints

具有能量收集设备和瞬时功率限制的智能电表隐私

• DOI: 10.1109/icc.2015.7249478
• 发表时间: 2015
• 作者: Giaconi G

Smart meter privacy based on adversarial hypothesis testing

基于对抗性假设检验的智能电表隐私

• DOI: 10.1109/isit.2017.8006633
• 发表时间: 2017
• 作者: Li Z

On Perfect Privacy

论完美隐私

• DOI: 10.1109/jsait.2021.3053432
• 发表时间: 2021
• 期刊: IEEE Journal on Selected Areas in Information Theory
• 作者: Rassouli B

Privacy Against a Hypothesis Testing Adversary

针对假设检验对手的隐私

• DOI: 10.1109/tifs.2018.2882343
• 发表时间: 2019
• 期刊: IEEE Transactions on Information Forensics and Security
• 影响因子: 6.8
• 作者: Li Z