The user side of sustainability: Modeling behavior and energy usage in the home

Society is becoming increasingly aware of the impact that our lifestyle choices make on energy usage and the environment. As a result, research attention is being directed toward green technology, environmentally-friendly building designs, and smart grids. This paper looks at the user side of sustainability. In particular, it looks at energy consumption in everyday home environments to examine the relationship between behavioral patterns and energy consumption. It first demonstrates how data mining techniques may be used to find patterns and anomalies in smart home-based energy data. Next, it describes a method to correlate home-based activities with electricity usage. Finally, it describes how this information could inform users about their personal energy consumption and to support activities in a more energy-efficient manner. These approaches are validated by using real energy data collected in a set of smart home testbeds.     

考虑智能用电新业务发展的电网评价方法

光伏云网、车联网、综合能源服务等智能用电新业务的快速发展会对电网运行产生新的影响。运用鱼骨图建立基于能量、信息、电能质量、有序充电及需求响应等多维度的智能用电新业务对电网运行影响评价指标体系,并从电网安全可靠性、协调平衡性、运行经济性及节能环保性方面评价智能用电新业务对电网的影响。同时,提出基于指标综合筛选的模糊综合评价方法。首先基于敏感度、贡献度、独立性对指标进行综合筛选,然后结合指标体系进行模糊综合评价,最后验证了指标体系与评价方法的实用性与有效性。     

基于门控循环单元的非均衡数据驱动异常用电检测方法

异常用电检测能够及时发现异常用电行为,在减少能源浪费和经济损失的同时能够维持安全、稳定的电网运行环境。智能电表的普及使得用电数据获取十分容易,为数据驱动的异常用电检测方法提供了充足的数据支持。然而,在实际应用过程中,异常数据较少导致的数据非均衡问题严重影响了模型的训练效果。因此,针对上述问题提出了一种针对非均衡数据的门控循环单元异常用电检测方法。该方法利用边界合成少数类过采样技术实现了对少数类数据的有效扩充。为了更好的捕捉用电数据的时序特征,采用了门控循环单元实现对用电数据的分类。为了验证该方法的有效性,基于非均衡数据集进行了对比实验。实验结果表明,该方法能够更好的数据扩充效果以及更准确的异常用电检测效果。     

基于云代理的智能手机3G通信端口节能模型

智能手机的无线通信端口是消耗电量的主要部件之一。随着移动互联网的普及,智能手机中主流的3G端口通信量不断增加,造成手机续航时间大大缩短。新兴的云计算服务平台采用云迁移技术对手机能耗进行优化,但针对通信能耗涉及较少。本文提出利用云代理技术实现智能手机中3G无线通信端口的节能模型。该模型主要针对占据总通信能耗较大比例的尾能耗进行优化,通过在云端采取尾能耗聚合与智能预取模型,实现3G端口能耗的降低。     

IntelliHome: An internet of things-based system for electrical energy saving in smart home environment

Despite there has been an increasing energy price due to factors such as supply, demand, government regulation, among others, users do not like to spend their time to analyze their power consumption and establish actions to save money. Hence, there is a need for smart solutions that help users to save energy at home in an easy way. The smart home concept is attracting the attention of both academia and industry to address this need. Nowadays, high volumes of data are available in the smart home context, facilitated by the growth of internet of things (IoT)-based devices and advanced sensing infrastructure. Therefore, it is necessary to automatically extract useful knowledge from this information to cost-effective use of energy at home. In this sense, this work presents IntelliHome, a smart-home system that aims to reduce electrical energy consumption at home. To this end, IntelliHome uses big data analytics technologies and Machine Learning and statistical techniques to provide users with a meaningful perspective of their electricity consumption habits aiming to actively involve them in the energy-saving process through real-time information and energy-saving recommendations. This work also discusses a case study and an evaluation aligned with the objectives of this work. The obtained results verify the effectiveness of the proposed system regarding electrical energy saving.     

Prediction of domestic power peak demand and consumption using supervised machine learning with smart meter dataset

Multimedia Tools and Applications - The prediction of electricity consumption is a vital foundation for smart energy management. Since the consumption of power varies with different appliances,...     

A Stackelberg game-theoretic approach to optimal real-time pricing for the smart grid

This paper proposes a Stackelberg game approach to maximize the profit of the electricity retailer (utility company) and minimize the payment bills of its customers. The electricity retailer determines the retail price through the proposed smart energy pricing scheme to optimally adjust the real-time pricing with the aim to maximize its profit. The price information is sent to the customers through a smart meter. According to the announced price, the customers can automatically manage the energy use of appliances in the households by the proposed optimal electricity consumption scheduling system with the aim to minimize their electricity bills. We model the interactions between the retailer and its electricity customers as a 1-leader, N-follower Stackelberg game. At the leader’s side, i.e., for the retailer, we adopt genetic algorithms to maximize its profit while at the followers’ side, i.e., for customers, we develop an analytical solution to the linear programming problem to minimize their bills. Simulation results show that the proposed approach is beneficial for both the customers and the retailer.     

一种基于时段特征的匹配算法在智能电表用电预测中的应用研究

用电需求预测问题对智能电网的稳定运行和用户体验的提升有着非常重要的影响。现有的工作大多是针对长、中期用电需求进行预测,而对短期内细粒度的用电需求的预测效果不佳。针对这一问题,基于特征匹配的思想,提出一种细粒度的用电需求预测算法。基于智能电表中记录的居民用电数据,提取每个用户的用电特征,并在智能电表中记录的实时用电数据上进行特征匹配,利用匹配到的特征进行用电量的预测。实验证明,该方法在细粒度的用电数据上取得了良好的性能。     

Privacy-aware electricity scheduling for home energy management system

In the context of smart grid, home energy management system (HEMS) needs to collect the fine-grained energy consumption data through smart meters. However, the fine-grained data contain the electricity consumption patterns of consumers, which can induce serious privacy issues. In order to protect the electric privacy of consumers, a privacy-aware electricity scheduling strategy for HEMS is proposed in this paper. Firstly, the basic scheduling model of HEMS is presented, and the basic scheduling objective is to minimize the electricity payment while satisfy the daily power demands of consumers. On this basis, a privacy-aware optimal scheduling model adopting rechargeable batteries is established, and the introduction of preference factor enables consumers to make a tradeoff between the total operation cost and privacy security. The electric privacy protection performance is measured by coefficient of determination and the number of features. Besides, the operation cost of batteries is also considered in the modeling process, and the influence battery capacity has on the performance of privacy protection is discussed. Simulation results show that the proposed method is effective and has strong practical application value.     

基于Bi-LSTM的家庭用电量预测

电网智能化升级改造将传统电网与先进的信息、智能技术相融合,实现电力行业的根本性变革。智能电表是智能电网系统中收集用户用电信息的代表性边缘设备,当前智能电表收集的用电量数据存在维度低、波动性强等特征,造成对未来用电情况难以预测的问题;同时对于未来边缘设备端用电量的预测,其他相关特征信息的不可得,此时研究基于单变量特征的用电量预测至关重要。为此,提出一种基于双向长短期循环记忆循环神经网络(Bi-directional Long Short-Term Memory,Bi-LSTM)的单变量家庭用电量预测模型,Bi-LSTM模型能够充分利用上下文的信息实现更准确的预测效果。通过西班牙某市真实的智能电表数据对提出的模型进行了验证,实验结果表明,该模型的预测性能相比传统LSTM、SVM方法有进一步的提高。     

智能电网互动化应用实时决策分析系统

在对配电和用电环节的信息化需求进行系统分析和研究基础上,实现了对终端智能电表和配网设备的实时监控和决策分析,通过智能电表等电子终端将用户和电网公司之间形成网络互动和即时连接,用信息流来反映能量流,实现电力数据读取的实时、高速、双向的应用,实现电力能量流和用户用电信息的实时可视化及决策分析。     

智能电网中基于MQTT协议的ABAC访问控制方案

在智能电网环境中,电力运营商和消费者通过智能电表进行大量高精度的用电数据的实时监测,用户机密数据持续暴露于未经授权的访问,在这种传统通信模式下,智能电表对家庭用户能源消耗的细粒度测量造成了严重的隐私安全问题,而现有的静态访问控制方法并不满足智能电网环境基于上下文的动态访问特性。针对此问题,提出一种基于物联网通信协议（MQTT协议）的访问控制方案,通过在MQTT协议中对树型结构的主题列表设计基于ABAC访问控制模型的动态上下文授权策略,并在WSO2系统使用XACML策略语言实现了提出的访问控制方案。性能评估结果表明,该方案能在较低的通信开销内支持动态的访问控制,以解决智能电网中用户的用电信息未经授权而泄露的隐私安全问题。     

Gridchain: an investigation of privacy for the future local distribution grid

As part of building the smart grid, there is a massive deployment of so-called smart meters that aggregate information and communicate with the back-end office, apart from measuring properties of the local network. Detailed measurements and communication of, e.g., consumption allows for remote billing, but also in finding problems in the distribution of power and overall to provide data to be used to plan future upgrades of the network. From a security perspective, a massive deployment of such Internet of Things (IoT) components increases the risk that some may be compromised or that collected data are used for privacy-sensitive inference of the consumption of households. In this paper, we investigate the privacy concerns regarding detailed readings of smart meters for billing purposes. We present Gridchain, a solution where households can opt-in to hide their consumption patterns and thus make Non-Intrusive Load Monitoring (NILM) more challenging. Households form groups where they can trade real consumption among themselves to achieve reported consumption that would be resistant to NILM. Gridchain is built on a publish/subscribe model and uses a permissioned blockchain to record any trades, meaning that dishonest households can be discovered and punished if they steal from other households in the group or the electricity company in the end. We implement and release a proof of concept of Gridchain and use public datasets to allow reproducibility. Our results show that even if an attacker has access to the reported electricity consumption of any member of a Gridchain group, this reported consumption is significantly far from the actual consumption to allow for a detailed fingerprint of the household activities.

     

基于协作频谱感知的智能电网通信网性能研究

将认知无线电频谱感知技术应用于智能电网的通信网中,可以有效提高频谱资源的利用率。现有研究仅考虑单用户单供电商,但是对需求响应管理性能与感知能耗权衡问题却没有给出理想的解决方案。建立基于多节点协作频谱感知的多用户单供电商智能电网通信网模型。在此基础上,为求解该模型需求响应管理和能耗感知性能权衡问题,提出基于多目标粒子群(MOPSO)的求解方法。仿真结果表明,所提协作频谱感知模型可以显著提高系统需求响应管理性能;MOPSO算法可实现系统需求响应管理性能和感知能耗的最佳权衡,有利于决策者根据实际要求灵活选择最优方案。     

基于时序分解的用电负荷分析与预测

在智能电网普及的大数据背景下,对电力数据进行精准的分析和预测对电网规划和经济部门的管理决策具有重要的指导意义,但大多数模型都只是在单一的时间尺度上进行研究。针对这一问题提出一种基于时序分解的后向传播算法的循环神经网络预测模型。通过对真实的居民用电消费数据以及外部因素数据统计处理,深入地分析了居民用电特点以及行为规律,并根据其数据的特征以及天气、节假日等外部因素对用户用电行为的影响建立预测模型,对用户未来时段的用电量进行预测。此外,考虑到居民用电消费数据的时序特征在不同时间尺度呈现不同的变化规律,通过时序分解建立预测模型来对用户用电行为的周期性和趋势性进行建模,并通过加权融合达到一起训练的效果,具有一定的协同性,提升预测精度。     

A hierarchical optimization model for energy data flow in smart grid power systems

Environmental concerns and high prices of fossil fuels increase the feasibility of using renewable energy sources in smart grid. Smart grid technologies are currently being developed to provide efficient and clean power systems. Communication in smart grid allows different components to collaborate and exchange information. Traditionally, the utility company uses a central management unit to schedule energy generation, distribution, and consumption. Using centralized management in a very large scale smart grid forms a single point of failure and leads to serious scalability issues in terms of information delivery and processing. In this paper, a three-level hierarchical optimization approach is proposed to solve scalability, computational overhead, and minimize daily electricity cost through maximizing the used percentage of renewable energy. At level one, a single home or a group of homes are combined to form an optimized power entity (OPE) that satisfies its load demand from its own renewable energy sources (RESs). At level two, a group of OPEs satisfies energy requirements of all OPEs within the group. At level three, excess in renewable energy from different groups along with the energy from the grid is used to fulfill unsatisfied demands and the remaining energy are sent to storage devices.     

Blockchain-empowered security and privacy protection technologies for smart grid

With the rapid development in society of the economy and of computational technologies, it is particularly important to build a secure, efficient and reliable smart grid architecture to provide users with high-quality electricity services. However, data collection and energy trading in public networks creates security and privacy challenges in smart grids. Blockchain technologies have the excellent characteristics of decentralization, immutability and traceability, which can resolve the security, integration and coordination problems faced by the traditional centralized networks for smart grids. The goal of this paper is to introduce and compare blockchain-based technologies in addressing the problems of privacy protection, identity authentication, data aggregation and electricity pricing for the data collection and power energy trading processes in smart grids. In addition, the existing challenges and future research directions of smart grids are discussed.     

Embedding intelligent eco-aware systems within everyday things to increase people’s energy awareness

There is a lack of energy consumption awareness in working spaces. People in their workplaces do not receive energy consumption feedback nor do they pay a monthly invoice to electricity providers. In order to enhance workers’ energy awareness, we have transformed everyday shared electrical appliances which are placed in common spaces (e.g. beamer projectors, coffee-makers, printers, screens, portable fans, kettles, and so on) into persuasive eco-aware everyday things. The proposed approach lets these appliances report their usage patterns to a Cloud-server where the data are transformed into time-series and then processed to obtain the appliances’ next-week usage forecast. Autoregressive integrated moving average model has been selected as the potentially most accurate method for processing such usage predictions when compared with the performance exhibited by three different configurations of Artificial neural networks. Our major contribution is the application of soft computing techniques to the field of sustainable persuasive technologies. Thus, consumption predictions are used to trigger timely persuasive interactions to help device users to operate the appliances as efficiently, energy-wise, as possible. Qualitative and quantitative results were gathered in a between-three-groups study related with the use of shared electrical coffee-makers at workplace. The goal of these studies was to assess the effectiveness of the proposed eco-aware design in a workplace environment in terms of energy saving and the degree of affiliation between people and the smart appliances to create a green-team relationship.     

基于用电大数据的用电异常状态辨识方法

用电异常状态的辨识是用电环节的重点和难点。本文基于计量自动化系统智能电能表所采集的用电大数据,对用电异常状态辨识方法进行研究。首先,基于用电海量数据及高维随机矩阵理论,研究分析了大维随机矩阵的协方差矩阵特征谱分布;然后,根据矩阵的统计特性提出基于用电大数据矩阵的用电异常状态辨识方法;最后,以贵州实际用电数据为例进行了仿真研究。仿真结果表明该文方法不仅能满足电网对可视性、时效性、可靠性、安全性的迫切要求,而且为数据驱动用电环节智能化、可视化监控提供了新思路。