Pitfalls of Machine Learning Methods in Smart Grids: A Legal Perspective

Reducing the power supply-demand gap and increasing reliability of power supply are the challenges of current energy management. Implementation of smart grid, smart meters and smart metering can be a possible solution for power demand reduction, efficient power supply management, and optimization of management resource usages. Smart meters include sophisticated measurement and calculation hardware, software, calibration and communication capabilities. For interoperability within a smart grid infrastructure, smart meters are designed to perform functions, and store and communicate data according to certain standards. In this work we discuss smart meter and various elements of smart metering, current state of the technologies related to smart grid, smart meter, advanced metering infrastructure (AMI), and meter data flow in smart grid. We also discuss standards related to smart meter, meter data format and data transmission, functions of smart meter, and functionalities of smart meters, currently deployed by utilities around the world.

Internet of Things connects people and things at any time and in any place. Smart grid is one of the promising applications of Internet of Things. Advanced metering infrastructure is one of the most significant components in smart grid which collects data from smart meters and sends the collected data to the utility center to be analyzed and stored. To implement two-way communication between smart grid components, machine-to-machine communications has a great potential. In this paper, we propose a machine-to-machine architecture for advanced metering infrastructure in Internet of Things-based smart grid and focus on finding the optimum number of collectors based on two metrics: collectors' cost and packet delay. We model advanced metering infrastructure as a queuing system and derive an expression for packet delay. We also propose a newperformance metric, the product of cost and delay (CD), so that by minimizing it, we can find the optimum number of collectors. The simulation results indicate correctness of our theoretical model and show that we can achieve a balance between cost and delay.

Public-key cryptography (PKC) is widely used in smart grid (SG) communications to reduce the overhead of key management. However, PKC comes with its own problems in terms of certificate management. Specifically, certificate revocation lists (CRLs) need to be maintained and distributed to the smart meters (SMs) in order to ensure security of the communications. The size of CRLs may grow over time and eventually may introduce additional delay, bandwidth, and storage overhead when various applications are run on SG. In this paper, we propose novel algorithms for creating customized CRLs with reduced size for IEEE 802.11s-based advanced metering infrastructure (AMI) networks. Rather than maintaining a huge-size single CRL that introduces unnecessary search time and storage, the idea is to cluster/group SMs within the AMI network and create CRLs based on these groups. The grouping is mainly done in such a way that they bring together the SMs that will be very likely to communicate so that the CRLs will be kept local to that group. To this end, we propose two novel grouping algorithms. The first algorithm is a bottom-up approach, which is based on the existing routes from the SMs to the gateway. Since the SMs will be sending their data to the gateway through the nodes on the route, this forms a natural grouping. The second approach is a top-down recursive approach, which considers the minimum spanning tree of the network and then divides it into smaller subtrees. Via grouping, the length of the CRL for each SM and the corresponding distribution overhead can be reduced significantly. Simulation results have shown that our approach can maintain a balance between the size of the CRL and the number of signatures generated by CAs while guaranteeing security of the communications.

Smart meters are key components of the smart grid infrastructure, connecting households to utility providers. With Internet connectivity to such devices, the smart grid Advanced Metering Infrastructure (AMI) is exposed to attacks by the adversary class. Any compromise of a smart meter may jeopardize the security of not just the household in question, but also the entire neighborhood network and possibly the utility provider's core network. One attack category that may pose a serious threat to the AMI is a meter compromise attack, wherein a smart meter is controlled by the adversary, with the intent of committing energy fraud. In this paper, we present a detection model based on a cumulative analysis of smart meter readings of clustered smart meters. The clustering of smart meters is done based on the physical deployment points of individual smart meters within a neighbourhood area network. Results obtained through simulation of the proposed scheme are promising.

The Advanced Metering Infrastructure (AMI) is a vital element in the current development of the smart grid. AMI technologies provide electric utilities with an effective way of continuous monitoring and remote control of smart grid components. However, owing to its increasing scale and cyber-physical nature, the AMI has been faced with security threats in both cyber and physical domains. This paper provides a comprehensive review of the crucial cyber-physical attacks and counter defense mechanisms in the AMI. First, two attack surfaces are surveyed in the AMI including the communication network and smart meters. The potential cyber-physical attacks are then reviewed for each attack surface. Next, the attack models and their cyber and physical impacts on the smart grid are studied for comparison. Counter defense mechanisms that help mitigate these security threats are discussed. Finally, several mathematical tools which may help in analysis and implementation of security solutions are summarized.

Research on data security of multicast transmission based on certificateless multi-recipient signcryption in AMI

The study examines the prospects and challenges of machine learning (ML) applications in academic forecasting. Predicting academic activities through machine learning algorithms presents an enhanced means to accurately forecast academic events, including the academic performances and the learning style of students. The use of machine learning algorithms such as K-nearest neighbor (KNN), random forest, bagging, artificial neural network (ANN), and Bayesian neural network (BNN) has potentials that are currently being applied in the education sector to predict future events. Many gaps in the traditional forecasting techniques have greatly been bridged by the use of artificial intelligence-based machine learning algorithms thereby aiding timely decision-making by education stakeholders. ML algorithms are deployed by educational institutions to predict students' learning behaviours and academic achievements, thereby giving them the opportunity to detect at-risk students early and then develop strategies to help them overcome their weaknesses. However, despite the benefits associated with the ML approach, there exist some limitations that could affect its correctness or deployment in forecasting academic events, e.g., proneness to errors, data acquisition, and time-consuming issues. Nonetheless, we suggest that machine learning remains one of the promising forecasting technologies with the power to enhance effective academic forecasting that would assist the education industry in planning and making better decisions to enrich the quality of education.

The delivery of energy and water meter data, management and control information on separate networks is expensive and defeats the gains of the Advanced Metering Infrastructure (AMI) Smart Grid (SG). In most cases, energy, gas and water services are offered by the same organizational entity, hence the use of different infrastructure for data, service delivery, control and management is expensive and highly illogical. There is a need for a combined energy and water infrastructure to reap the benefits of the AMI SG. Furthermore, combined metering will result in accurate billing, potential cost savings, and improved resource management. This work therefore develops and investigates a combined energy and water AMI smart metering framework. This is possible through a thorough understanding of the AMI technological standards. The implementation of such a system is not trivial, as it depends on many factors: environmental, geographical, technological, economical, regulatory and the existing legacy infrastructure. Optimal technological implementation choices are developed towards an integrated AMI infrastructure. An experimental test bed is developed for delivering energy and water metering data to the utility. The optimal placement results favor the system of separating energy and water actuators at the home area network of the SG while using an integrated communication system. Such a system is feasible, given the different evolution of electricity and water meters and their placement at the home area network, and enables water metering to benefit from the more advanced electrical metering infrastructure.

This study evaluates the feasibility of utilizing the 2100 MHz frequency band for the implementation of a Narrowband Internet of Things (NB-IoT) network in the Purwokerto region using two schemes, In-Band and Guard-Band. This research aims to provide recommendations for the most efficient scheme to support the implementation of Smart Metering based on Advanced Metering Infrastructure (AMI) in the area. The method used is simulation based on radio planning software, considering the number of existing sites, network capacity, and coverage. The key performance parameters analyzed include Reference Signal Received Power (RSRP), Signal to Interference plus Noise Ratio (SINR), and throughput. The simulation results show that both schemes can be used for the implementation of the NB-IoT network in Purwokerto, but the Guard-Band scheme provides better performance with RSRP of –94.39 dBm, SINR of 6.61 dB, and throughput of 14,947 bps, compared to the In-Band scheme, which results in RSRP of –99.39 dBm, SINR of 6.61 dB, and throughput of 14,623 bps. Based on these results, the Guard-Band scheme is recommended for implementation because it provides superior performance in the RSRP and throughput parameters, which are crucial for supporting the quality of Smart Metering AMI services. The contribution of this research is to provide new insights into the implementation of NB-IoT in Purwokerto and compare two frequency allocation schemes within the 2100 MHz band. Keywords: NB-IoT, Smart Metering, In-Band, Guard-Band, 2100 MHz.

The protection of smart meters (SMs) from cyberattacks is of utmost importance because SMs in advanced metering infrastructure (AMI) are physically unprotected and produce a large amount of sensitive data. Due to scalability, the SMs are small-sized and low-cost devices having low computational capabilities. The algorithms that are designed to complete the security requirements of SMs should be lightweight. To address this issue, this paper proposes a lightweight security solution to address the man-in-the-middle attack, data tempering, and blockchain-based data provenance. Received signal strength indicator (RSSI) is used to generate link fingerprints, which are used along with pseudo-random nonce to secure AMI. The proposed algorithm detects the involvement of adversarial node or meter tempering by computing other values along with 0 and 1 as the average of consecutive RSSI and difference between the RSSI of connected static SMs. Pearson correlation coefficient ( $\rho $ ) of 0.9102 is achieved when no adversarial node is present in between the connected SMs having mobility in one or both SMs. Negative or approximately equal to zero values of $\rho $ are computed when the adversary is present in the AMI or any of the SM in the AMI is forged. For blockchain-based data provenance, all the hash values of the packet header are 100% matched with the hash functions present at the data concentrator unit (DCU), which shows no adversary’s involvement in AMI. For cases when the adversary is in the AMI, hash functions show no match with the hash values present at the DCU.

In the modern world, there are numerous opportunities that help in the detection of electricity theft happening in the realm of electricity grids due to the widespread shifting of people from old metering infrastructure to advanced metering infrastructure (AMI). It is done by studying the consumers’ energy consumption (EC) readings provided by the smart meters (SM). The literature introduces a variety of machine learning (ML) and deep learning (DL) strategies to use EC data for identifying power theft in smart grids (SGs). However, the existing schemes provide low performance in terms of electricity theft detection (ETD) due to the usage of imbalanced data and using schemes in an individual manner. Moreover, the existing detectors are validated using a limited number of performance evaluation measures, which are not suitable for conducting model’s comprehensive validation. To tackle the above mentioned problems, an ML boosting classifiers based stacking ensemble model (MLBCSM) is proposed followed by adaptive synthetic sampling technique (ADASYN) in the underlying work. Data preprocessing followed by data balancing and classification are the three major parts of the model introduced in this work. Besides, the EC data acquired from the consumers’ SMs is used for detecting electricity theft. Moreover, the simulation results reveal that MLBCSM combines the benefits of adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), histogram boosting (HistBoost), categorical boosting (CatBoost), and light gradient boosting (LGBoost). Additionally, the model’s validation is ensured via different metrics. It is deduced via extensive simulations that the proposed model’s outcomes are superior to those produced by the individual models in terms of ETD.

Internet of Things connects anythings at any time in any places. Smart grid is one of the promising applications of Internet of Things. Advanced metering infrastructure is one of the most important components of smart grid which aggregates data from smart meters and sends the collected data to the utility center to be analyzed and stored. In this paper, we have proposed an aggregation-based indirect architecture for advanced metering infrastructure architecture in smart grid Internet of Things and focused on finding the optimum number of regional aggregators in its neighborhood area network (NAN) based on three metrics: regional aggregator (RA) power consumption, cost of RAs and wide-area aggregator, and NAN lifetime. We have developed a power consumption model for each RA and define NAN lifetime based on that. We also defined a new performance metric, the product of power consumption and cost (PC), so that by minimizing it, we can obtain the optimum number of RAs. The simulation results indicate correctness of our theoretical model and show that we can achieve a balance between NAN lifetime and RAs' cost.

A review of non-technical loss attack models and detection methods in the smart grid

Advanced Metering Infrastructure (AMI) enables two-way communication between smart devices and utility control centers. This involves remote monitoring and control of energy consumption as well as other parameters in the electrical power network in real time. However, increasing technologies in AMI due to huge deployment of smart meters, integration of devices and application of sensors, demand a strong architectural model with the best network topology to guarantee efficient usage of network resources with minimal latency. In this work, a resource efficient multi-hop network architecture is proposed using hybrid media access protocols. The architecture combines queuing and random-access protocol to achieve optimal network performance. Numerical results show that the probability of delay incurred by an arbitrary smart meter depends on the mean and distribution of the queue switch over a period. It is also observed that for a single queued system, the throughput performance is equal to the existing hybrid method. As the number of smart meters increases to 500, the throughput of the proposed method improves by 10% compared to the existing method. Likewise, as the number of smart meters increases to 500, the delay reduced by 15% compared to the existing method. Keywords: Advanced Metering Infrastructure; hybrid media access protocols; Smart Meter; Smart Grid; Power Network.

The detection and identification of Distributed Denial-of-Service (DDoS) attacks remains a challenge in cloud/edge/fog computing environments. It usually requires network middleboxes, such as deep packet inspectors (DPI), for detection task mostly. But clouds and fogs have native powerful telemetry systems that are not yet fully exploited for DDoS detection; and provide so much information that could aid attack identification tasks as well. Machine Learning (ML) algorithms can help one diving into the richness of cloud’s native data collection services, which have a multitude of metrics from both physical and virtual hosts. This paper evaluates the use of ML algorithms over datasets collected from a experimental testbed based on OpenStack. Controlled attack scenarios were used to investigate the ability of ML for tasks such as detecting and identifying SYN_Flood and GET_Flood DDoS attacks mixed, in different proportions, with legitimate clients. kNN and Random Forest ML algorithms were trained and tested, and for evaluation the metrics accuracy, recall, precision, and F1-score were used. Our experiments presented about 87% of accuracy in the detection of SYN_Flood and GET_Flood DDoS attacks, whereas Snort IDS mostly fails to detect the latter attack by processing the corresponding packet traces. Also, the detection of PING_Flood DDoS attack was tested without training as an initial evaluation towards the generalization of the proposal.