Explore the vast range of titles available.

Simple analysis of energy consumption patterns recorded by smart meters can be used to deduce household occupancy. With access to higher-resolution smart-meter readings, we can infer more detailed information about the household including the use of individual electric appliances through non-intrusive load monitoring techniques. The extent of privacy concerns caused by smart meters has proved to an obstacle in the roll-out of smart meters in some countries. This highlights the need for investigating smart-meter privacy. Mechanisms for ensuring smart-meter privacy fall in broad categories of data manipulation, demand shaping, and load scheduling. In smart-meter data manipulation, the smart meter collects real, potentially high-resolution data about the energy consumption within the house. This data is then manipulated before communication with to utility providers and retailers. The manipulation could be non-stochastic, such as aggregation, binning, and down-sampling, or stochastic, such as additive noise. In demand shaping and load scheduling, smart-meter readings are communicated without any interference but the consumption is manipulated by renewable energy sources, batteries, or shifting loads to render non-intrusive load monitoring ineffective. In this study, the author reviews these approaches and presents several methods relying on homomorphic encryption, differential privacy, information theory, and statistics for ensuring privacy.

To detect energy theft attacks in advanced metering infrastructure (AMI), we propose a detection method based on principal component analysis (PCA) approximation. PCA approximation is introduced by dimensionality reduction of high dimensional AMI data and the authors extract the underlying consumption trends of a consumer that repeat on a daily or weekly basis. AMI data is reconstructed using principal components and used for computing relative entropy. In the proposed method, relative entropy is used to measure the similarity between two probability distributions derived from reconstructed consumption dataset. When energy theft attacks are injected into AMI, the probability distribution of energy consumption will deviate from the historical consumption, so leading to a larger relative entropy. The proposed detection method is tested under different attack scenarios using real-smart-meter data. Test results show that the proposed method can detect theft attacks with high detection percentage.

Smart-meter technology advancements have resulted in the generation of massive volumes of information introducing new opportunities for energy services and data-driven business models. One such service is non-intrusive load monitoring (NILM). NILM is a process to break down the electricity consumption on an appliance level by analyzing the total aggregated data measurements monitored from a single point. Most prominent existing solutions use deep learning techniques resulting in models with millions of parameters and a high computational burden. Some of these solutions use the turn-on transient response of the target appliance to calculate its energy consumption, while others require the total operation cycle. In the latter case, disaggregation is performed either with delay (in the order of minutes) or only for past events. In this paper, a real-time NILM system is proposed. The scope of the proposed NILM algorithm is to detect the turning-on of a target appliance by processing the measured active power transient response and estimate its consumption in real-time. The proposed system consists of three main blocks, i.e., an event detection algorithm, a convolutional neural network classifier and a power estimation algorithm. Experimental results reveal that the proposed system can achieve promising results in real-time, presenting high computational and memory efficiency.

Addressing data anomalies (e.g., garbage data, outliers, redundant data, and missing data) plays a vital role in performing accurate analytics (billing, forecasting, load profiling, etc.) on smart homes’ energy consumption data. From the literature, it has been identified that the data imputation with machine learning (ML)-based single-classifier approaches are used to address data quality issues. However, these approaches are not effective to address the hidden issues of smart home energy consumption data due to the presence of a variety of anomalies. Hence, this paper proposes ML-based ensemble classifiers using random forest (RF), support vector machine (SVM), decision tree (DT), naive Bayes, K-nearest neighbor, and neural networks to handle all the possible anomalies in smart home energy consumption data. The proposed approach initially identifies all anomalies and removes them, and then imputes this removed/missing information. The entire implementation consists of four parts. Part 1 presents anomaly detection and removal, part 2 presents data imputation, part 3 presents single-classifier approaches, and part 4 presents ensemble classifiers approaches. To assess the classifiers’ performance, various metrics, namely, accuracy, precision, recall/sensitivity, specificity, and F1 score are computed. From these metrics, it is identified that the ensemble classifier “RF+SVM+DT” has shown superior performance over the conventional single classifiers as well the other ensemble classifiers for anomaly handling.

The steadily increasing usage of smart meters generates a valuable amount of high-resolution data about the individual energy consumption and production of local energy systems. Private households install more and more photovoltaic systems, battery storage and big consumers like heat pumps. Thus, our vision is to augment these collected smart meter time series of a complete system (e.g., a city, town or complex institutions like airports) with simulatively added previously named components. We, therefore, propose a novel digital twin of such an energy system based solely on a complete set of smart meter data including additional building data. Based on the additional geospatial data, the twin is intended to represent the addition of the abovementioned components as realistically as possible. Outputs of the twin can be used as a decision support for either system operators where to strengthen the system or for individual households where and how to install photovoltaic systems and batteries. Meanwhile, the first local energy system operators had such smart meter data of almost all residential consumers for several years. We acquire those of an exemplary operator and discuss a case study presenting some features of our digital twin and highlighting the value of the combination of smart meter and geospatial data.

Rapid evolution of electricity distribution networks challenges the maintenance of up-to-date information in electricity utility databases. This hinders the ability of utilities to understand phase connectivity and topology of users in their distribution networks. Extensive research has been conducted to develop smart meter data-driven phase identification and topology determination approaches as alternatives to the conventional, time-consuming, and expensive approach of manual inspection. However, the majority of such approaches are challenged by low levels of smart meter penetration in distribution networks, entailing further investigation. The objective of this paper is to contribute to this challenge by proposing an alternative smart meter data-driven approach of user phase identification, verification, and topology determination and testing the method on a real Australian distribution network under varied levels of smart meter penetration. This paper first presents a smart meter data-driven user phase identification tool using k-means clustering. Then, a smart meter data-driven user phase verification and topology determination approach is introduced by analyzing voltage-to-power sensitivities obtained from linear regression. Four distinct linear regression models are developed and compared to recognize relevant parameters and input variables leading to the most reliable sensitivities. The overall process proposed in this study demonstrated high accuracy at original smart meter penetration of 75% of the case study DN. The performance at reduced smart meter penetrations of 50% and 25% is also examined and discussed in the paper.

Energy theft refers to the intentional and illegal usage of electricity by various means. A number of studies have been conducted on energy theft detection in the advanced metering infrastructure using machine learning methods. However, applying machine learning for energy theft detection has a problem in that it is difficult to obtain enough electricity theft data to train a machine learning model. In this paper, we propose a method based on anomaly pattern detection to detect electricity theft in data streams generated from smart meters. The proposed method requires only normal energy consumption data to train the model. Previous usage records of customers being monitored are not needed for energy theft detection. This characteristic makes the proposed method applicable in real situations. Experiments were conducted using real smart meter data and artificial attack data, including the preprocessing of daily consumption vectors by standard normalization, the construction of an outlier detection model on normal electricity consumption data of randomly chosen customers, and the application of anomaly pattern detection on test data streams. Some promising results were obtained, notably, that attacks of types 4, 5, 6 were detected with an average F1 value of 0.93 and average delay of 19 days.

Time-series smart meter data can record precisely electricity consumption behaviors of every consumer in the smart grid system. A better understanding of consumption behaviors and an effective consumer categorization based on the similarity of these behaviors can be helpful for flexible demand management and effective energy control. In this paper, we propose a hybrid machine learning model including both unsupervised clustering and supervised classification for categorizing consumers based on the similarity of their typical electricity consumption behaviors. Unsupervised clustering algorithm is used to extract the typical electricity consumption behaviors and perform fuzzy consumer categorization, followed by a proposed novel algorithm to identify distinct consumer categories and their consumption characteristics. Supervised classification algorithm is used to classify new consumers and evaluate the validity of the identified categories. The proposed model is applied to a real dataset of U.S. non-residential consumers collected by smart meters over one year. The results indicate that large or special institutions usually have their distinct consumption characteristics while others such as some medium and small institutions or similar building types may have the same characteristics. Moreover, the comparison results with other methods show the improved performance of the proposed model in terms of category identification and classifying accuracy.

Renewable Energy Communities (RECs) are crucial in advancing decentralised and sustainable energy systems. Accurate forecasting of short-term electricity demand is essential to support REC operational planning, improve energy efficiency and reduce dependence on external supply. This study presents a practical and effective forecasting framework for hourly electric load in a residential REC in Loureiro, Portugal. The approach is based exclusively on exogenous variables like weather conditions and calendar-based features. Two tree-based machine learning models, Extreme Gradient Boosting (XGBoost) and Categorical Boosting (CatBoost), were applied to predict REC electric energy demand. The performance of the models was evaluated using standard regression metrics, including the Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and the coefficient of determination (R2). Both models achieved high accuracy on unseen data, with CatBoost yielding significantly better results: MAE, RMSE, and R2 are equal to 1.21 kWh, 2.31 kWh, and 0.9718, respectively. An intraday 24-hour comparison confirmed the model’s ability to capture intraday consumption dynamics. These findings highlight the potential of exogenous feature-driven forecasting approaches to support efficient energy management in communityscale systems.

One of the main challenges for city development is to ensure a sustainable water resource management for the water supply system. A clear identification of the urban water consumption patterns supports policy and decision makers in managing the water resources, satisfying the total demand and, at the same time, reducing losses and identifying potential leakages or other issues in the distribution network. High resolution smart meters have widely shown to be an efficient tool to measure in-pipe water consumption. The collected data can be used to identify water demand patterns at different temporal and spatial scales, reaching the end-uses level. Water consumption patterns at building level can be influenced by multiple factors, such as socio-demographic aspects, seasonality, and house characteristics. The presence of a garden that requires summer irrigation strongly alters the daily consumption pattern. In this framework, we present an innovative approach to automatically detect the presence of garden irrigation, identifying daily average water consumption patterns with and without it. The proposed methodology was tested in a residential area in Northen Italy, where 23 smart meters recorded data at 1-minute resolution for two years. Results show very good performances in distinguishing between days with and without garden irrigation. The derived average normalized water consumption patterns for both scenarios can help decision makers and water managers to regulate the pressure regimes in the distribution network correctly.HighlightsHigh resolution smart meter data have been used to identify water consumption patterns;Automatic detection criteria to classify days with and without garden irrigation are designed;Normalized average water consumption patterns with and without irrigation are proposed.