Analysis on Secured Cryptography Models with Robust Authentication and Routing Models in Smart Grid

2.1 Routing models

Zhang and Leung [1] improved load flow routing and V2G scheduling for regulatory service delivery using a hierarchical system model is proposed. Firstly, the author introduced PFRs into the power system to formulate the problem of OPF routing at the grid level. Using the semidefinite programming (SDP) relaxation, the author turned the original NP-hard issue into a convex problem, which can be solved in linear time. Das and Bera [2] discussed the novel advancement of ZigBee-based tunable clustering scale-free neighbourhood area networking (TSFNAN) for establishing smart grid connectivity. Using the hop minimization feature of scale-free and customizable clustering techniques, the complexity of ZigBee-based multihop mesh networks between gateways and smart metres (SMs) is lowered.

Critical information should be routed to safe pathways in the event of smart grid cyber-physical coupling failures, which is expected [3]. Equality of information flow is a priority in power communication networks. Information flow is critical, but the value it provides depends on the current condition of the power system.

By analysing the information flow's cyber-physical sensitivity, the model presented by Xu et al. [3] can be used to calculate its value. A CPS-RRM (CPS-RRM) with a priority mechanism that takes into account cyber-physical disturbances is the next phase. For the Big-M solution of the reformulated CPS-RRM, an updated version of the CCG method is used instead of the standard one.

The dynamic nature of SGCNs makes it difficult to pinpoint the best way for reliably transmitting data. When trusted routing is used, no matter how logically constructed the rule set and membership function were in earlier fuzzy logic systems, they waste computational memory and reduce node energy efficiency. When Velusamy and Pugalendhi [4] used the water cycle algorithm (WCA), they developed an approach for automatically fine-tuning rule sets and membership functions for decision variables. WCA-based exploitation, evaporation, and rainfall can be used to discover the near-optimal rules and locations for membership functions. Network Simulator-2 is used to evaluate the SGCN routing algorithm in an experimental setup (NS2). The suggested model is put through its paces in three distinct scenarios, each with its own set of tests: Each can be evaluated on its own; the relationship between them can be evaluated separately; and the relationship between them can be evaluated as a whole.

A smart grid's power system and communications network are tightly intertwined. A malfunction in one network might have a domino effect on other networks, causing even more problems. A model was created by Kong [5] to evaluate the risk of communication channel failure and to quantify the impact of this failure in terms of lost load. We devised an optimization approach to lessen the impact of an initial route failure. A NP-hard method is required to calculate how much power is being wasted. Dynamic programming is further transformed by using Bellman's optimality equation to solve the problem. A Gauss-Seidel value iteration strategy, which the author offer, can be used to address the dynamic programming problem.

2.2 Security models

As the Smart Grid is created, the safety of the communication network must be considered. Hostile attacks can be conducted at any time since the wireless network is open and unpredictable. Using Bayesian inference and the Dempster-Shafer (D-S) theory, Velusamy et al. [6] created a new framework for measuring direct and indirect trust. The node's trustworthiness in the analytical hierarchy process is determined by cross-layer attributes such as rate of transmission, absorption rate, and signal strength (AHP). Fuzzy theory combines fairness when computing link trust and reliable routing using cross-layer metrics. To test the proposed fuzzy-based trust routing approach, extensive testing were carried out with malicious nodes.

The high cost of phase shift measurement equipment has made it a popular study topic for some time now. A system cannot be made visible solely by Phasor Measurement Unit (PMUs). An appropriate communication network has to be in place to transfer all PMU data [7]. In order to ensure full observability of a power system, Zhu et al. [7] proposed a model in which OPLP examines the appropriate placement of PMUs and communication links(CLs). Each CL's communication capability, as well as the locations of PMUs and CLs, is captured in the OPLP issue. This ensures that PMU data is transmitted reliably and on time.

There is a direct link between a cyber-physical system and the electrical grid. Because of their dependence on interconnected systems, smart grids are at risk of failures piling on top of one another. Power grid failures could be caused by a breakdown in the communication network, and vice versa. Keeping an initial failure from spreading can stop a cascading network disaster.

A sufficient number of electricity communication lines between power nodes as well as the control centre is required to accomplish this. The number of power-disjoint pathways can be used as a metric of rigidity. Reducing reliance between the electrical grid and communication networks is the model proposed by Kong [8] ensuring that systems are as resilient as possible. The author determined which power source powers which communication node and which communication line connects which power source to the control centre by using this type of relationship. To solve a Maxflow problem generated from this connection Menger's Theorem is applied.

The smart grid uses two-way communication and extensive features to promote sustainability and efficiency, but these features also raise serious privacy concerns for users. Individual privacy is protected in smart grid systems via cryptographic techniques like encryption algorithms, which can only provide limited and simple functionality. According to Menger's theorem, the size of a minimal cut set in a finite graph is equal to the greatest number of disconnected paths between any two vertices. In order to analyse the maximum flow problem, we assume that (1) all arc capacities are integers and (2) whenever the defined as the sharing arc I j), the network also has arc I j) (j, i). Since we permit arcs with zero capacity, the second possibility is lenient. Because of the limited resources of smart metres, these solutions need the use of complex asymmetric cryptography. Li et al. [9] describedSecGrid, an SGX-enabled smart grid system. Grid utilities employing the system can process confidential customer data swiftly and securely because of the usage of SGX-certified technology. Only the smart metres require AES encryption, with SecGrid's well-designed security procedures. To verify the models design's superiority, security analyses and experiments are conducted.

A key part of the smart grid is data aggregation that protects user privacy while simultaneously expanding the applications of data aggregation and meeting the demands of fine-grained data analysis. Customers may worry about their privacy as a result of the fact that classic multidimensional data-aggregation systems are vulnerable to coalition attacks from the gateway and control centre, which rely on an established authority. A decentralised decryption system based on ElGamal homomorphic cryptography is presented by Zuo et al. [10], which can withstand the coalition assault from both the Gateway (GW) and the Control Center (CC). It is also important to note that the proposed method does not rely on an authority that is not totally trustworthy in the real world. According to a thorough security analysis, the proposed technology is capable of meeting the security criteria of the smart grid.

2.3 Cryptography models

Secure communication between smart metres and the smart grid is only possible if an authenticated key protocol can be developed, which has recently received substantial attention. In most circumstances, a mutual authentication method can be reached without the involvement of a trusted third-party. Elliptic Curve Qu-Vanstone (ECQV) implicit certificates were used by Qi and Chen [11] to construct an authenticated key agreement method for the smart grid and achieve some significant advancements in this sector. There is no need for any pairing in order to assure mutual authentication scheme agreement and strong credential privacy with minimum computation and communication costs in the suggested approach.

Smart IoT devices can now be linked in new ecosystems to improve energy systems’ efficiency and reliability by managing various energy sources. The amount of data generated by smart grid IoT devices necessitates the use of a cloud server. Because the cloud server’s data is accessible to many people, it must be authentic and confidential. Using proxy re-encryption, a third party can decrypt an encrypted file and re-encrypt it without viewing the original content. This is the best choice for this type of communication. There are currently no smart grid encryption systems that do not demand a significant amount of bandwidth and computing time. Hussain et al. [12] offer an IoT device certificate-based signcryption with proxy re-encryption (CBSRE) to reduce communication and computation expenses. The suggested CBSRE approach relies on a hyperelliptic curve cryptosystem with minimal parameters and an 80-bit key size to provide security and efficiency.

For resource-constrained smart metres, Garg et al. [13] proposed a lightweight and safe authentication approach that provides security, anonymity and mutual authentication. In order to create a key agreement procedure that can be verified by both parties, elliptic curve cryptography and one-way hashing algorithms are employed. In addition, it gives the capacity to establish and verify trust and understanding between SMs and neighbour area networks. Because these organisations communicate over an insecure network, smart metering relies on it. Long-term testing has proven that this protocol is more secure than the industry’s existing standard while also consuming less communication and compute resources.

The new phase of electronic power networks is represented by the terms of smart grid. Electricity is tracked by sensors, communications, and control devices as it travels from the source to the end user. For clients who want to know exactly how much energy they’re using at any one time, a growing number of smart metres are being placed across the globe. With less network traffic and processing resources, increased energy efficiency, and accuracy of data, privacy in industrial ecosystems may have to be given more concern. Computer resources, communication overhead, and hiring a trustworthy third party have all been used to assault the privacy of customers in various researchers. Ali et al. [14] proposed a symmetric encryption approach for industrial ecosystems in the IoT context. The proposed model exhibits better encryption accuracy rate.

It’s essential that the smart grid be protected against cyberattacks. SG’s security demands necessitate the proper implementation of key establishment processes. In recent years, Singapore has seen a spate of diverse proposals for large new skyscrapers. There are, however, only two smart metres that are capable of ensuring the highest level of privacy for their users. In the smart grid, utilities and customers share data and electricity in a two-way exchange. A smart meter’s data can be sent to service providers (SPs) and also control signals in the other direction. Users can better utilise electricity resources by evaluating the present condition of production and supply through this two-way interaction. Abbasinezhad-Mood et al. [15] proposed an efficient and secure key agreement technique to address the issue of key escrow, while avoiding the security issues of earlier anonymous schemes.

2.4 Malicious user detection models

Analyzing and making use of massive amounts of metre data can be beneficial to decision-makers. As a result, the processing of data from several metres has received a great deal of attention in recent years. Even so, it has the potential to reveal personal information about its users. Data processing methods for smart grids that are effective and private are presented by Shen et al. [16]. Using Horner’s Rule and homomorphic encryption, the proposed technique keeps metre data private while randomly permuting it on a huge scale. Without understanding where the data comes from, analysis centres can simultaneously execute a variety of operations such as variance, comparison, and linear regression analysis. The proposed technology is capable of ensuring data security and source anonymity, according to the results of an investigation. In terms of computation and communication, the proposed protocol is shown to be cost-effective.

The employment of cutting-edge technology and equipment in the smart grid provides substantial advantages over traditional electrical networks. There are a slew of new security issues brought on by the smart grid’s new hardware and software. Malicious individuals aiming to steal electricity can gain access to smart metres at any time and from any location. This has the effect of making it more difficult to catch people who steal electrical equipment. User authentication analysis is one way researchers have used to track down attackers. Due to either a lack of precision or prohibitive expenses for installing monitoring devices, these solutions are not ideal. A limited number of monitoring devices will be used to locate malicious users as fast as feasible. Investigations of power theft begin with a thorough review of prior records of electricity theft and discrepancies between reported and projected normal consumptions. Based on these findings, a suspicion honest assessment inspection (SAI) method has been devised by Xia et al. [17], in which the most suspected users are first investigated. The other users will be investigated using a binary tree-based examination approach. The user’s hunches guide the creation of the binary tree. Binary tree node inspection order is also influenced by the suspicions.

To regulate the flow of real power, power grid phase shifters are utilised by Chakrabartyand Sikdar [18]. Due to contractual obligations, transmission lines might become overcrowded and cross-network power flows can become unpredictable. These phase shift directions are communicated via Supervisory control and data acquisition (SCADA) networks in a smart/automated grid. Consequently, it is vulnerable to cyberattacks, especially those conducted in secret. It is possible for malicious phase shift directives to disrupt important transmission lines by interfering with cross-network trading. Although this control system is vital, little attention has been paid to cyber-attacks against it. An algorithm or approach to detect various attacks, including some that are undetectable, is proposed for the first time in this research. The suggested algorithm is based on the current-to-terminal-voltage ratios of branches or nodes. These indexes have been demonstrated to be effective in the detecting context through the use of mathematical models. With these indices, it was found that the proposed technique worked well with phase shifters in the IEEE 118-bus system and was computationally light and simple to implement.

A real-time non-probabilistic assurance can be used to detect load redistribution (LR) assaults on the smart grid that intend to overflow. A reliable and complex detection method is essential in the event that the standard bad-data detectors fail to detect LR attacks. Kaviani and Hedman [19] proposed a detection strategy based on a fundamental knowledge of the physics of the electric grid. It is able to uncover a structure beneath the surface that may be abused by an attacker to solve the root of the problem. The most successful attack paths and the most vulnerable transmission assets are then identified, and an efficient method is provided. The ideal attack and sensitive buses revealed in this study were used to develop an index that may be used in practise with minimal disruption.

Using a monitoring method called state estimation (SE), smart power grids will play an important role in the future of smart cities. Edge computing-based Internet of Things (IoT) is used to estimate the state of the imbalanced home distribution grid in the region. Future control actions depend on the SE results, but the accuracy of the data collected by the distributed measuring devices is as important concept. Because of this, the SE module is at risk from attacks based on manipulated data. At this time, the most attention is being paid to an attack known as fake data injection (FDI), which has the potential to impair routine network operations while remaining undetected. Tranet et al. [20] proposed a nonlinear physical constraint model for a stealthy attack on FDI systems. For testing purposes, an IEEE 13-node test feeder and WSCC 9-bus system are employed.

2.5 Attack detection models

The Smart Grid Monitoring System (SGMS) is a critical component in ensuring the safety of the smart grid. Due to the overwhelming volume of notifications generated by SGMS, executives are usually left perplexed. When it comes to handling alarms and extracting attack events, smart grids confront significant difficulties. It is impossible to apply most existing security event analysis methods to the power grid because of its high reliability and limited attack tolerance. SGMS warnings could be utilised to detect attacks, according to this research. Using IP correlation, Zhang et al. [21] generated an approach that generates an alert graph, which is then aggregated into a prospective attack chain. On the other hand, the number of early attack chains is reduced by negative causal correlations and non-cascading events.

Cyber-physical (CP) attacks, vulnerabilities and mitigation approaches for the smart grid applications are examined by Amin et al. [22]. Due to cyber risks, the performance of the smart grid is critically impacted by the rapid development of physical technologies in power electronics for connecting renewable energy sources that include cyber frameworks. There is a risk of major cyber-attacks because of the use of electronic equipment connected to communication networks in smart grid applications If this happens, one approach is to physically isolate digital controllers. Cyber-physical systems demand additional attention and security for power electronic systems in the smart grid because of CPSs.

A data integrity attack (DIA) is one sort of cyber-attack that puts the IoT-based smart grid at risk. As long as the attacker has complete or insufficient knowledge of the system's architecture and branch parameters, it is impossible to detect and undermine smart grid state estimate with the highly synthesised DIA. The branch parameters can't be easily accessed or deduced by an attacker in practise. They can change or be disrupted at any point in time. Zhang et al. [23] provided the zero-parameter-information DIA (ZDIA) so that the attacker can carry out stealthy data manipulation attacks without knowing anything about the branch parameters. The cut line's geometry is all that is required for this assault type. If an attacker has access to all the buses in a group of one-degree super-buses that are only connected to the outside by a single cut line, users can arbitrary modify the state estimations of all of them.

A vital role for ICT is played in today's smart cities in controlling demand response management. In a smart grid environment, sensors are used extensively to keep track of everything. These devices are used to alert the central control station when a high-tension power supply line is malfunctioning. Due to their location on high-tension electrical lines in an open environment, sensors may be vulnerable to both physical and cyberattacks. As a final resort, a hacker could pretend to be a sensor, gateway, or control centre. Cyber and physical attacks on a smart grid must be averted in order to maintain a secure and effective monitoring system. Inspired by these considerations, Badar et al. [24] proposed an identity-based authentication method for power supply line surveillance in a smart grid. When tested thoroughly, the protocol designed is impervious to both physical and cyber-attacks on sensors. Reduced calculation needs make this protocol more cost-effective than other relevant ones.

Over time, concerns about power grid cybersecurity have increased as communication technologies have become more intertwined. State Estimator (SE) must be secured against cyberattacks because of its critical role in the control of electricity grid. False Data Injection Attacks (FDIA), a subtype of assaults targeting this module, have received a lot of attention. A model for detecting FDIA was proposed by Jorjani et al. [25]. It is impossible to get back to the original values of grid variables that have been tampered with. Using an adaptive optimization method, grid elements that were targeted inside the attack can be recovered while making as few changes as possible to those that were not. It has been suggested that the recovery algorithm's effectiveness be measured using the Recovery Quality Index (RQI).

Developing an AMI is essential to implement smart grids. Reliability and security of the smart grid are directly impacted by the security of AMI's digital infrastructure. Analytical models established by Zhang et al. [26] are used to test the AMI system's resilience against Distributed Denial-of-Service (DDoS) assaults. In order to create a model of how the state of an AMI communication network varies over time, this research aims to create and describe a dynamic differential system for the network as a whole. An attack on AMI's dynamic differential system is examined, and a defence technique is proposed that best distributes AMI system security resources to minimise defensive losses and costs.

The traditional routing models performs mitigation of the negative effects of both proactive and reactive routing systems. There is no delay caused by establishing a route while communicating across short distances. Decreased overhead caused by reactive routing for distant destinations is observed.