Using the internet of things in smart energy systems and networks

Highlights

• IoT offers wide range of applications in the energy sector, like energy generation, renewable energy integration, etc.

• We have conducted a comprehensive and critical IoT study on smart energy systems and networks.

• IoT in smart energy applications; IoT in data transmission networks; and IoT in energy production resources are reviewed.

• With critical thinking and clear vision, many new solutions for smart energy systems are provided.

• IoT global energy market exceeded USD 6.8 billion in 2015 and is projected to reach USD 26.5 billion by 2023.

Abstract

Private businesses and policymakers are accelerating the deployment and advancement of smart grid technology innovations that can support smart energy systems. Technological advances such as the Internet of Things (IoT) provide a broad range of energy sector applications, such as transmission and distribution, energy supply, power generation, renewable energy integration, load demand management, etc. We have conducted a comprehensive and critical IoT study on business applications and smart energy systems for this objective. Four main areas of smart energy systems have been chosen, including: i) the use of IoT in business; (ii) the use of IoT in smart energy applications; (iii) the use of IoT in data transmission networks; and (iv) the use of IoT in power generation and terminal hardware/devices. Each research area is further divided into different sub-areas; for example, IoT business includes energy operational policy, energy management and planning, energy management systems, business models, and customer services. Energy forecasting, state monitoring and estimation, anomaly detection, data mining and visualization are among the IoT applications in smart energy systems. Cloud computing, edge computing, and quantum computing are provided using IoT in data transmission networks. A variety of renewable sources, pricing, and load management strategies involve the use of IoT in energy generation. Many new solutions for smart energy systems are provided with critical thinking and clear vision, and key industries for IoT revenue generation and application development are described. This study aimed to provide a clear insight into IoT devices' recent developments in smart energy systems, supported by high-quality published literature. It is noted that in 2015, the IoT worldwide energy market exceeded USD 6.8 billion and is projected to reach USD 26.5 billion by 2023, with a compound annual growth rate of 15.5 % in 2016−23.

Introduction

The IoT is a new paradigm for smart energy systems. The insights derived from new IoT-connected devices are used to build new technologies, increase performance and productivity, address critical issues, improve decision-making in real time and create creative and fresh experiences. However, as more devices connect, power utilities face increasing interoperability, fragmentation, and security challenges. Many information technologies (ITs) provide integrated, scalable software and hardware technologies specifically designed to address diverse market needs that effectively fit existing energy networks to improve reliability, security, and performance. Interoperability and interconnectivity are key features in designing advanced power management systems for industrial sites (Wei, Hong, & Alam, 2016). The history and timeline of IoT are based on four industrial revolutions. The first industrial revolution involves the production of steam and water. In the second industrial revolution of IoT, belt and mass production are being conveyed. In the third industrial revolution, information and technology and electronic goods were significantly improved. The fourth IoT industrial revolution is currently underway, and most work is being done on users of cyber-physical systems for different types of energy systems.

Implementation of IoT in different industries and sectors has been extensively discussed in the literature (Da Xu, He, & Li, 2014; Talari et al., 2017). These studies conclude that many companies shift from visual controls and age-related maintenance schedules to remote monitoring, IoT network design, and predictive maintenance. Companies need to evolve and adapt to ensure that the resources and assets required are securely linked and make sense of the massive amount of data they generate. And they've got to do this cost-effectively and quickly.

How often does IoT have an impact on the energy industry? The IoT is expected to reorganize the energy market, from controlling the electricity grid to generating electricity and increasing energy efficiency. Data science and sensor-based technology are making solar and wind farms more automated and efficient. The use of IoT is not limited to an area of the energy system, but includes smart energy buildings (Pan et al., 2015; Pérez-Lombard, Ortiz, & Pout, 2008; O’Dwyer, Pan, Charlesworth, Butler, & Shah, 2020), energy monitoring (Hassan, Chou, & Hassan, 2019), energy harvesting (Ko & Pack, 2019), energy use and optimization schedule (Ding & Wu, 2019), energy internet (Lin, Deng, Liu, & Chen, 2017), efficient data collection (Orsino et al., 2016), energy integration (Zakeri, Syri, & Rinne, 2014), industries and smart homes (Hui, Sherratt, & Sánchez, 2017; Biljana & Kire, 2017; Lee, Hsiao, Huang, & Chou, 2016; Dae-Man & Jae-Hyun, 2010), smart grid applications (Hui et al., 2020; Batista, Melício, Matias, & Catalão, 2013), energy sustainability (Khatua et al., 2020), auto-energy management (Zouinkhi, Ayadi, Val, Boussaid, & Abdelkrim, 2020), and so on. These studies conclude that the IoT has a great potential to provide long-term data on the overall condition of the power generation plant, which significantly improves plant efficiency. Actual-time information is used to fine-tune plant operations, maximize conversion of fuel and reduce maintenance costs.

The IoT promotes modern technology. IoT technology enables all energy consumption and production components to be connected, improves operational visibility, and provides real leverage at every stage of energy flow from use to supply and end-user. This collaboration strengthens the economies of both hydrocarbon and renewable energy production. IoT strategies for the power industry help companies minimize operating costs and maintenance through human effort and system optimization process. IoT and energy technology make it possible to optimize the operation of wind farms, maximize operations and significantly reduce costs. Fig. 1¹ visualized the IoT hype cycle plot for IoT. It explains that seven technologies underpin IoT, including: data virtualization tools, IoT integration, IoT edge architecture, low-cost development boards, IoT services, machine learning, data management (MDM), or energy-consumption data are used in smart energy systems.

The use of IoT can be classified into three main classes, including: i) use of IoT software in the energy sector; ii) use of IoT applications in the energy sector; and iii) use of IoT to the end-use industry. IoT in the energy software sector includes energy analytics software, data management software, real-time streaming analytics, remote monitoring software, and IoT security software for energy operations. Applications for IoT in the energy sector include IoT in power distribution, equipment monitoring, mobile workforce management, field monitoring, and IoT in energy management. The use of IoT in the industrial sector includes the fuel sector (oil, coal), the power sector and IoT in the gas and oil sectors. By enabling suppliers to assess useful information regularly, IoT technologies offer significant cost savings and efficiency. Automation and simplification will improve customer support and business processes.

Based on the importance and available literature, we conducted a comprehensive and up-to-date study of IoT in smart energy systems used in business applications and networks. This study was divided into four main components: IoT business models, IoT applications, IoT networks, and IoT in different energy environments. Business models include the operational policy of energy management systems, control systems, energy planning, construction, and network upgrades. Various types of IoT applications have been reviewed, such as energy prediction, data mining and machine learning (Table 1 shows the summary of machine learning models), anomaly detection, state monitoring and data visualization. Three database networks are cloud computing, edge computing, and quantum computing, and their use in energy systems is briefly investigated. Finally, the use of IoT in grid station, renewable energy sources, load demand management, and price control of end-user is briefly discussed. The recent advances in IoT for the energy sector and the measures that companies should take to make more extensive use of IoTs are discussed. More authenticated references are used to provide vital support to literature. Many new solutions have been reported with critical thinking.

The rest of the paper is divided into seven sections. Section 2 explains the use of IoT in different sectors, such as the use of IoT in businesses, the use of IoT in various energy applications, the use of IoT transmission networks, and the use of IoT in the utility environment. These sections have been classified in a further sub-section with additional useful information. Section 3 described IoT's capacity and capabilities, such as sensing and actuation devices, localization, user interface, corporate and communications, addressability, processing, and information identification. Section 4 identifies the leading sectors for IoT revenue generation and application development, and Section 5 presents the recent progress in IoTs. Section 6 proposes that the companies' measures to adopt IoT, and Section 7 concludes this study.

IoT's technological challenges based on energy systems include: i) fault tolerance and discovery; ii) interoperability and complexity of the software; iii) power supply and stability; and iv) data volume and interpretation. Four challenges IoT faces in the future are described below.

Fault tolerance applies to network operations under sensitive conditions, such as node failure or network resource depletion (Sharma, Song, You, & Atiquzzaman, 2017). Fault-tolerance is used to network operations under sensitive conditions, such as node failure or network resource depletion. Three key aspects are challenging, such as the operating properties of fault tolerance, including power reliability, node reliability and route reliability. If all of these properties are addressed together, IoT networks can be made effective. IoT-based devices' discovery challenges include the automation of route tagging and identification management operations centers, integration/discovery and on-demand services (Patel, Patel, Scholar, & Salazar, 2016; Mainetti, Patrono, & Vilei, 2011).

Recent studies on IoT device interoperability have focused on the selection of protocol gateways. However, protocol appears to be pushing the interoperability issue rather than resolving it, so there are optimization problems with an increased number of servers, reconfiguring effort, requiring overhead processing and bandwidth (Derhamy, Eliasson, & Delsing, 2017; Benhamou & Estrin, 1983; Bandyopadhyay & Sen, 2011; Noura, Atiquzzaman, & Gaedke, 2019). Scalability to such an ecosystem is a major challenge, but it is crucial to a useful IoT. The industry demands that the protocol bridge and its interoperability be scalable, transparent, verifiable and secure. It must also allow for the provision of verbose coverage and promote the quality of service.

There are three major IoT-based software complexity devices, including rectification complexity, overall complexity, and input and output complexity (Niu et al., 2020; Kim, Nam, Park, Scott-Hayward, & Shin, 2019). The overall complexity metrics include the number of functions, the number of code lines, the code lines defining variables and procedures, the circles' complexity, the complexity of some of the major mathematical algorithms, and the total recursive network call layers. Output and input complexity include metrics such as response parameters used by features, heaps and function call stacks and function variables. Rectification and intuitive complexity are the number of annotated sections of the code. These are the major challenges to the complexity of IoT software designed for use in the energy sector. In addition, quality indicators and software complexity lead to vulnerability research in three key areas of research: improving and adjusting vulnerability tracking performance, repairing and evaluating vulnerability strategies, and cost-control and predicting vulnerability recurrence. Table 2 lists the major IoT-focused software, hardware and computing companies. It's just a partial list, but more focused on IoT-focused economic activity list is given in reference (Bank, 2017).

Business security systems, both software and hardware, do not share or process large quantities of detailed information from connected devices and sensors (Ande, Adebisi, Hammoudeh, & Saleem, 2020). It is challenging to enrich machine-generated data and quickly integrate data from business applications such as Marketo and Salesforce and other data repositories or centers. Moreover, today's companies need to pursue solutions that effectively enable data to communicate with each other to leverage the company's full information.

The power supply is a unique challenge for IoT's self-contained networks, particularly when meeting the "continuously on" requirements. The device is wireless only, even if it lasts for a battery. IoT devices rely on energy harvesting or batteries, or day and night power, as shown in Fig. 2² . In addition, engineers must be careful when designing an IoT device to reduce power usage (Measurement, n.d.).

Improving power grids using IoT devices is challenging (G.Radhakrishnan, 2020; Leu, Chen, & Hsu, 2014). Ensuring the protection, reliability, durability, and stability of internet applications and services for energy systems is essential for maintaining the internet's support and use. The problem is how to keep the devices stable. Many IoT devices, such as consumer goods and sensors are developed globally, including conventional network-connected devices. The possible number of interconnected connections between these applications is unparalleled. In addition, many of these applications can interact and create dynamic and unpredictable links to other systems on their own. In addition, current IoT-related resources, approaches and techniques require new attention to address these challenges (McEwen & Cassimally, 2013; Bedi, Venayagamoorthy, Singh, Brooks, & Wang, 2018; He et al., 2016).

Data volume is a major challenge for IoT devices as they are overloaded with large data volumes. There is a wider range of connected devices and no-one-fits-all of these solutions. The flow of data streams must be a challenge both inside and outside the device. This is particularly challenging in the power sector and industrial organization, where large industrial consumers and manufacturers usually collect trillions of data sets from various devices, internal business applications, and sensors (Abdul Rahman, Daud, & Mohamad, 2016; Perera, Ranjan, Wang, Khan, & Zomaya, 2015). Big data volume includes a variety of other challenges, including volume (e.g., data sizes from petabytes to zettabytes), velocity (e.g., data rate and speed), variety (e.g., data formats such as unstructured, structured, and semi-structured), veracity (e.g., reliability and quality), variability (e.g., data flow at different rates) and values (e.g., data information). In addition, the development of IoT devices capable of managing accuracy, learning complex patterns, training complexity and transfer learning is challenging (Amanullah et al., 2020). Any user-to-environment interaction will involve proper visualization software, which will also illustrate the interpretation and sensing mechanism of data scenarios through security, privacy and data management is a challenging task (Singh, Tripathi, & Jara, 2014; Jayavardhana, Rajkumar, Slaven, & Marimuthu, 2013).

Other challenges that are essential and must be considered and addressed as a matter of priority include outdated software and hardware, use of default and weekly credentials, ransomware and malware functions, predicting and preventing hacker attacks, known or identified affected/disabled devices, security and data protection issues, automation of vehicle security, home privacy, etc.

Although technologists describe IoT as a further step towards a better society, academics and social analysts have a number of questions and concerns about the pursuit of the ever-present technological revolution. They conclude that there are major technological impacts, such as IoTs, decision-making, privacy, autonomy, and human agency. Another concern is that the use of IoT is being built quickly without sufficient recognition of the deeper security issues involved and the technological changes needed. When IoT expands, cyber-attacks or hacker attacks can become an extreme physical threat. The marketplace for publicly available sensor data may serve commercial and security interests as well as assist criminals and spies in the detection of high-value targets. IoT devices can measure the vulnerability of any cyber-attacks. Major disputes over the use of IoT devices, including walled-out internet (for example, cross-border attacks, regulatory divergence, and economic protectionism), security-based AI issues (e.g., gird control, power system integration and system synchronization), cloud attacks (e.g., cybersecurity, big data), botnet issues (e.g., critical infrastructure, somewhat dumb devices, and public infrastructure (e.g., a huge amount of data handling capability, complex models, experts and expenses, durability, machine learning complexity, weather impacts, security encryption, situational awareness and locking of networking).