A survey on new trends of digital twin technology for power systems

Abstract

Recently, the Digital Twin (DT) technology, which joints the physical environment and virtual space, has drawn more attention in industry and research academic plans. In general, the virtual model representations of the physical objects are created in the DT manner to simulates the characteristics and behaviors of the real-word system. Applying a supervisory system not only can reduce the failures of components, but also preserve the overall costs associated with the system at a minimum. This paper reviews the DT applications in the power system, while its advantages in wind turbines, solar panels, power electronic converter, and shipboard electrical system will be briefly discussed. The potential benefits of contemporary technologies to ameliorate the DT in the industry are studied. Besides, it provides a great technique to assess and analyze system performance. As a basis for DT, various new emerging developments as an example of artificial intelligence (AI), big data, the internet of things (IoT), and 5 G are reviewed.

Keywords

Index Terms: Digital Twin (DT)Artificial Intelligence (AI)ship power system big data Internet of Things (IoT)

1 Introduction

Nowadays, the digital twin has achieved a strong impact as a modern technology that can revolutionize manufacturing outlook [1, 2]. This technology conducts as a replica of reality offers a method for simulating, forecasting, and optimizing physical systems and operations [3]. Furthermore, merging with an intelligent algorithm, systems obtain progress in supervisory and optimization activity [4], advance novel products and all types of services [5], also expanding business structure and production [6].

Digital technology that allows the unification of components into the shop floor is called Industry 4.0. It has been identified generally as one of the most effective response of companies to financial hardship [7]. Also, it has been established in the Internet of Things (IoT) [8] that recommended embedding electronics and connection of network into devices (i.e., “things”) to allow exchange and accumulation of information toward the internet [9]. Therefore, devices can be direct appropriately from long distances and create a new path among virtual and physical systems. Besides, a cyber-physical system (CPS) is used as an embedded system that operates at a virtual and physical system communicating with physical devices and sensors and react to the real world [10 –12].

These days, the use of big data is becoming commonplace for enterprises to outperform their peers. Established rivals and new entrants can use the tactics emerging from the evaluated data to perform, evolve and capture demand throughout most industries respectively. Big Data lets businesses generate potential avenues for innovation and whole new types of firms that can integrate and interpret data from the market. These businesses provide sufficient knowledge that can be collected and evaluated about goods and services, customers and vendors, customer desires. Every business uses data in its own way; the more successfully a business uses its knowledge, the more potential it has to expand. The organization should take information from any source and examine it to find answers that would allow: Cost savings: As vast volumes of data need to be processed, such Big Data technologies such as Hadoop and Cloud-Based Analytics may offer cost benefits to business and these tools can help to find more successful ways of doing business. The high speed of tools such as Hadoop and in-memory analytics will quickly detect new data sources that enable organizations to instantly analyze data and make fast decisions based on the lessons learned [13].

The three main reasons allowing for digital twin development are:

IoT systems are able to capture a huge amount of information simply as well as send it in almost real-time to a digital twin.

Digital knowledge allows us to gain more in-depth insight into the physical asset data.

Machine learning algorithms can evaluate data collecting and make forecasts, optimize digital twin based on information gathered, and calibrate the general model and its specifics.

This article aims to investigate issues related to developing a digital twin application specifically for the power system and presents methods for implementations in the wind turbine, power converter, solar panels, and shipboard electrical systems. This paper studies the literature review after the introduction part, which reviews the digital twin’s structure and base form. Then, using the digital twin in the power system will be investigated deeply. The contribution of the survey is stated as follows:

in this work, an attempt is accomplished to emphasize the applicability of the digital twin concept in various practical case studies such as wind turbine plant, shipboard systems, power electronic converters from systematical point of view.

for cost-effective and quick deployment of DT, the key enabling IoT technologies like reliable measurement components, low-cost information storage, and Big Data analytics have been introduced.

the development of advanced control methodologies (e.g., machine learning, heuristic algorithms) in the DT framework for various case studies have been elaborated.

The remainder of this survey has been arranged as follows; Section II reviews the digital twin’s basic concept. The application of DT in the power system has been elaborated in Section III. Moreover, DT’s potential in the industry of wind turbines has been discussed in Section IV. Section V gives a survey on the deployment of DT along with the control strategies in the context of maritime systems. The integration of DT and Big data for the photovoltaic systems have been analyzed in Section VI. A practical application of the DT on power electronic systems presented in Section VII. The drawbacks and challenges related to the utilization of DT technology have been emphasized in Section VIII. Lastly, the conclusion of this paper has been demonstrated in Section IX.

2 Fundamental concept of digital twin

Industry 4.0 brings a new path to real-time supervising and synchronizing real-world activities to the virtual environment because of virtual-physical connectivity as well as CPS elements’ networking. The digital twin is a digital or simulative replica of physical assets or products. It was first introduced by Grieves in 2002 [14]. NASA was one of the first to use this technology mission. By gathering real-time data from the installed sensors, DT is able to connect the virtual and real world. In fact, the collected data is either locally decentralized or centrally stored in a cloud. Then, the data is simulated and assessed in a virtual copy of the assets. The parameters are applied to real assets, after receiving the information from the simulation. This integration of data in real and virtual representations helps in optimizing the performance of real assets [15, 16]. DTs are employed in different industries such as utilities, healthcare, automotive, manufacturing, and construction. They are the next big thing in the fourth industrial revolution of new product’s evolvement and processes. Also, it can help to predict future events by analyzing the data. Figure 1 shows the reflection on the outlook of smart manufacturing, in which DT can have an impact on future manufacturing for assets and people [17].

Fig. 1

Overall view of smart manufacturing.

The asset is connected to cyberspace by using the digital twins. Producers acquire a vision of real-world performance to analyze received data and make a decision to improve the operation. It also links workers on the shop floor. Personal information like emotional status, height, weight is formed to predict the working conditions of people in a factory. Besides, it supports machine-human cooperation in order to boost the workforce’s health and harvest the maximum production rate [18, 19].

The truth is that digital twins, if the method is easy, will generate value without machine learning and AI. For example, if there are limited variables and a readily discoverable linear relationship between inputs and outputs, then there would be no need for data science. The vast majority of applications, though, have multiple variables and multiple data sources and need data processing expertise to make sense of what’s happening. By applying machine learning to every manufacturing process, the process can become smarter by gaining more detailed data and forecasts, as well as understanding visual and unstructured data. people not only open up opportunities to find previously unknown trends in your data by incorporating machine learning into your workflow, but also create a single learning framework that can handle complicated data.

Using three parts are necessary to improve the digital twin: i) data models are needed in order to obtain the physical system’s specifications ii) a connection procedure that conveys bi-directional information among DT and its physical elements iii) data operation module which is able to draw out data of various multi-source to build an exact model of the physical system. Figure 2 represents a reference model of the digital twin. For the proper cooperation of these three elements, they need to work continuously.

Fig. 2

Structure of digital twin for power system [24].

Actual sensor data could be used by the digital twin to practically replicate locations and physical operating environments on computers. Therefore, the operation should be assessed previously to mitigate the source of error. Also, it leads to less maintenance and wear of the system. These digital twins are becoming increasingly important in more dynamic industries and can chart the lifespan of goods, processes, or services.

A digital twin is a full and operational virtual representation of an asset, subsystem, or device that blends digital aspects of how equipment is designed with real-time aspects of how it is managed and maintained (PLM data, concept models, manufacturing data). Simulation, diagnostics, prediction, and other specialized use cases are made possible by the ability to link to data stored in various locations from one shared digital twin directory. It consists of several main parts which will be elaborated on in the following paragraphs.

Design: During the design process, modeling and analysis will be used to check and inspect the final 3-D design and make sure that all components match together. As well as interrelationships between these elements, simulations involve mechanical, thermal, and electrical.

System integration: 3-D system-level visualizations could validate restrictions including spatial footprint and physical relations.

Interactions, including data transmission and control capabilities, as well as mechanical and electrical activity and what-if scenarios, can be replicated by interacting with the digital twins of other components. Integration effort on site and the customer’s related downtime was minimized.

Diagnostics: For example, in 3-D visualization, an inspection of the digital twin will help troubleshooting. Virtual-reality glasses may have an overlay on the physical equipment allowing field technicians to simulate parameters. Non-observable data, such as temperatures of non-accessible sections or material tension, may be applied through simulations.

Prediction: In conjunction with predictive algorithms, historical and current operating and sensor data offers insights into the state of infrastructure and the probability of various types of failure. It helps schedule fair repairs and eliminates unplanned downtime.

Advanced services: provided that the advanced service parameters (IoT connectivity, analytics algorithms, etc) are preconfigured in the digital twin, they can be enabled when the equipment is installed and the customer subscribes to these services. In the optimal case, no further engineering is required.

During design and setup, the digital twin embraces and reduces ambiguity when promoting applications and maintaining availability and stability through condition tracking and specialized facilities. The freedom to refer to data stored in different locations from one standard digital twin directory is the secret to the varied facets of a digital twin. Through accessing data from elsewhere in the product lifecycle, a digital twin is able to provide the data needed for multiple usage cases.

A reliable data model that includes the calculated variables and the mapping that contributes to a digital representation is necessary for the digital twin. This purpose demands that measurements conveniently meet the digital twin representation, which, in turn, requires a connectivity infrastructure to be engineered.

3 Power system application

Renewable energies encourage countries to achieve their worthwhile development goals by ensuring access to renewable, safe, efficient, and inexpensive electricity [20 –23]. One of the most significant accomplishments of modern-day infrastructure is the power grid. Power supply enriches our daily lives. It is enabled by wires, poles, substations, and so on. However, there is a vast amount of digital data, allowing utilities to schedule, operate, and support the grids with a real-world model computer. With developments such as decentralization and recycling, monitoring and sharing of these data are getting more dynamic. Only minor data mistakes may have significant consequences. Utilities choose technologies that weave together their digital data should be prepared for a prosperous digital future. The digital twin can be used in numerous applications in power systems. Wind turbine, power converter, solar panel, and shipboard electrical system are among important power system which focuses on them and tries to analyze the operation. Figure 2 demonstrates the structure of the DT for the power system. The connection between the physical grids, the power system, and the controllers are identical to the relationship between the remaining vehicles on earth, the space module during flights. It allows the operators to figure out the physical system under challenging circumstances and makes a credible decision [24].

Digital innovations are changing the power generation market, bringing new insights into the planning, management and maintenance of power plants, and uncovering new opportunities to enhance system efficiency and maximize the availability and reliability of equipment. Most of the data produced within the power industry was historically confined to remote monitoring and diagnostic facilities, where turbineactivities were monitored for monitoring. Many of the offerings of power generation in the digital environment have been around the following main areas:

Increasing Accessibility. This strategy directs managers of power plants to optimize system checks, control routine maintenance practices, and optimize condition-based maintenance by balancing reliability and inventory management of operations and maintenance activities.

Improvement of Results. In terms of versatility, this approach handles plant processes and enhances efficiency to increase sustainability in the ever-changing environment.

Field Support Remote. This automated approach and virtualization allows for early warning of irregularities and potential disruption to the tracking and diagnostics of power plant facilities.

Danger and Conformity. The convergence of OT and IT networks has placed greater emphasis on the protection of power plant cyberattacks. The global landscape protection umbrella tracks networks for increasing cyber threats and decreases risk while maintaining compliance with industry standards.

Control of Fleets. This monitoring and diagnostics framework offers total accountability across machine, facility, and fleet levels, including asset diagnostics, efficiency monitoring, and optimization of operations to build new ways for utilities to control and sell their assets.

As a physical power electronic converter system contains elements of random (stochastic) uncertainty during operation, a deterministic (non-random) real-time model to use as the digital twin of the system (and its subsystems) cannot consider these elements that affect the behavior of the system. Examples of these stochastic and uncertain elements that can influence power converter behavior include electromagnetic interference (EMI), thermal effects, component tolerances and manufacturing defects, fluctuating loads and sources, and more. By not considering these uncertainties, a deterministic digital twin may not accurately model the behavior of a physical system during real operation. While periodically refreshing and training the digital twin model definitions from measurements of the converter in normal operation can assist with keeping the twin matching the converter in terms of expected values, a deterministic model still cannot predict behavior influenced by random elements with probable outcomes. Implementing the probabilistic models for the controller embeddable digital twins is a possible solution to address the failures.

Probabilistic models are used in the controller-embedded DTs of the proposed diagnostics approach to take into account the uncertainty that naturally occur in the modeling phase due to the limited knowledge of the plant parameters (manufacturing defects, component value under tolerances), and to take in account stochastic environmental aspects that are difficult to predict or measure but affect the power converter operation (electromagnetic interference, thermal effects, sensor noise). Two common probabilistic modeling methods include Monte Carlo (MC; sampling method) and generalized Polynomial Chaos Expansion (PCE; analytical method [25 –30].

4 Wind turbine system

At the beginning of creating wind turbines, there was more emphasis on how the turbines generated energy rather than on the lifetime [31]. Nevertheless, because of the number, scale, and position of the wind turbines, the handle of these systems is physically challenging. Utilizing a virtual physical turbine model enable remote turbine operation, as data is transmitted to the controller. Moreover, it helps maintenance processes to be simplified, load management and power delivery, and disturbances and faults detected and anticipated.

Generally, renewable energy, and especially wind energy, is growing in size and capacity and is expected to continue to expand [32, 33]. In theory, wind turbines convert wind into mechanical power to kinetic energy, then, again into electrical power. This occurs to the main shaft with a pushing force on the tips, raising and rotating. The drivetrain is mounted in the wind turbine’s nacelle, where the electricity is transmitted. Then the electricity from the blades flows to a generator, which has an electrical grid connection [34]. The idea of using wind energy to generate electrical power emerged from the beginning of the 20th century. Also, the horizontal axis of the wind turbine was the most commonly deployed turbine [35]. There has been substantial demand growth since the 2015 Global Wind Survey [36].

General types of wind turbines have been classified according to two parts: wind turbines with variable-speed (VS-WT), and also, fixed-speed wind turbines (FS-WT) [37]. Several advantages like better power management and increased recovery of energy, as well as a decrease in transient load, could be accomplished by comparison of VS-WT and FS-WT [38]. An excellent control strategy has a vital impact on the behavior and characteristics of the WT [39, 40].

Two major functional areas could be categorized for VS-WT, high and low wind speed of the turbine, depending on the measured wind speed. The main objective of the controller at low wind velocity is to maximize the wind energy’s capturing by reducing the turbine parts’ uncertainties. Also, the most important goal at the above-noted wind speed is to sustain WT’s strength [41].

The stable function of a wind turbine system is highly dependent on WT’s control system in various areas of operation. In order to limit the absorbed power at the above-rated wind speed region by the wind turbine system, numerous type of traditional and modern control structures have been presented for designing effective pitch angle controllers such as proportional-integral (PI) with new variations [42, 43], fuzzy logic structure [44, 45], linear parameter-varying (LPV) [46], the nonlinear control structure [47], optimum control strategy [48], robust control [49], sliding mode control (SMC) [50] and model predictive control (MPC) [51]. Traditional PID controller and also its derivative is rarely capable of achieving great function in disturbance areas, as new high-order of WTs, multivariable, and high nonlinearity of couple. However, for a nonlinear high-order system, fuzzy logic control that typically employs Logical analysis to regulate gains of the PID, provided in [52].

The structure of the Wind turbine consists of a two-mass model is shown in Fig. 3, widely deployed in many works, in order to explain the dynamics of the WT. As for the following equation, the direct relation with wind speed the total wind power exists [51].

Fig. 3

Two-mass variable speed wind turbine model [41].

$P_{w} = \frac{1}{2} ρ {AV}^{3}$ (1)

In which A is equal to the swept area of the turbine (m²) and ρ is equal to air density (kg/m³) and V is equal to wind speed (m/s).

The turbine would absorb the total wind energy If the wind speed after passing the turbine was zero. Although, because of the loss of wind, coveying all energy is practically impossible. So, the coefficient (C_p) of output power is obtained to represent the wind turbine’s aerodynamic efficiency. Employing C_p, turbine aerodynamic power (P_a) can be expressed in the following terms: $P_{a} = \frac{1}{2} ρ C_{p} {AV}^{3}$ (2)

The (C_p) parameter is defined as a nonlinear function which is based on pivotal elements: blade pitch angle (β) and tip speed ratio (λ) which is expressed in the following equation:

$C_{p} (λ, β) = 0.5176 (\frac{116}{λ_{i}} - 0.4 β - 5) e^{- 21 / λ_{i}} + 0.0068$ (3)

The variables λ_i could be expressed below: $\frac{1}{λ_{i}} = \frac{1}{λ + 0.08 β} - \frac{0.035}{β^{3} + 1}$ (4)

Also, λ is estimated using the speed of wind for the upstream of the rotor and the blade tip and as follows: $λ = \frac{R ω_{r}}{V}$ (5) where w_r is equal to the velocity of rotor angular. the wind turbine’s power coefficient curves are shown in Fig. 4 as a function of β and λ.

Fig. 4

power coefficient C_p (λ, β) curves for the wind turbine.

Nowadays, the deployment of WT power has been raised dramatically, and a further rise has been predicted, and expansion of the offshore wind turbine industry was projected. Most of this is expected to be mounted in Great Britain, Germany, and China [53, 54]. Though there are advantages to offshore wind turbines, it also faces some difficulties. The added hydrodynamic forces and connection to the grid are not only a concern but also wear, as the turbines and their systems are in direct contact with water. Moreover, there are problems with installation and maintenance because of bad weather and timetable [55]. Operation and maintenance are defined as a vital aspect of a WT budget, which consists of about 25 percent of the total cost of producing electricity, or about 75 percent to 90 percent of the investment cost of a 750 kW turbine [56, 57]. Offshore operating and repair charges are evaluated up to four times higher compared to onshore costs because of being laborious to access areas and entry facilities as well as the extreme expense of the required expert personnel. Therefore, using an optimized repair and service system could be a substantial opportunity for cost savings [58].

Figure 5 shows the critical parts of offshore and onshore WTs in terms of downtime. Sometimes, drivetrain components contribute to the most downtime of WTs. The various components of the WT as well as its associated breakdown and downtime levels are considered in [59, 60]. Between 1993 and 2006, Landwirtschaftskammer Schleswig-Holstein (LWK) obtained failure data from over 650 wind turbines [61]. Between 1997 and 2005, a Swedish study was carried out on wind turbine failures. The gearbox appears in the wind turbine as a highly rated breakdown and downtime feature [62]. Also, bearings are considered to be an essential part of failure [63]. Bearing failures are triggered by cases related to lubrication or power [64]. The breakdown for blades, gearbox, and engine is considerably greater than the electrical sub-assemblies [65]. Vibration monitoring is the most common security technique for spinning equipment. This technique is used not only on bearings and towers but also on gearbox, rotor, and blades [66].

Fig. 5

Critical segments of offshore and onshore WTs in terms of downtime [58].

Energy is absorbed by the bearings and tower which support radial and axial forces and gain the specific vibrating of the gearbox. Also, failure of the gearbox occurs at the bearing since being exposed to fatigue at high risk [67]. Monitoring is achieved for wind turbines by measuring the vibration at the gearbox and rotor wheels and bearings. The primary bearing, in particular, is known as it greatly affects the quality of the other bearings [68].

A digital twin could be used for the gas and oil companies of offshore WTs [69]. Less exposure to more advanced structures raises management problems and is one of the latest issues faced by digital twins. Not only are digital twins used for production and operation, but they may also be employed to link back-end process systems such as accounting and human resources to accomplish specific process results in supply chain operations [70]. Further development of the technology requires to be accomplished for achieving a centralized, automated controlled maintenance system, solving existing problems of automated maintenance [71]. It could be operated autonomously for the WT or additional remote and complicated structures, allowed to imagine for the DT technology, which executes maintenaning duties. Because of the small weather windows open for the entry that should consider highly useful for offshore WTs and the robots will be permanently mounted. Maintenance robots with remote control are commonly employed in the nuclear industry [72].

Step-up transformers are widely set up as dry-type transformers, particularly in offshore wind farms. Their performance at offshore wind turbines is excellent because of their low flammability and resistance to moisture. However, considering the different wind and weather conditions in marine habitats, their thermal-electrical loss must be noticed. Also, dry-type transformers in offshore WTs are taking into account highly thermal strained position, i.e., the winding hotspot. Moreover, lifetime of transformer l’s calculations can be extended for diagnostic and prognostic testing purposes within the context of digital twins. The study of transformers degradation is established upon a standard load profile of offshore WT’s which involves the effects of many characteristics for the operating conditions and transformer [73].

The power switches are among the important parts experiencing different levels of the short and medium-term thermal cycles, particularly in offshore WTs. The DT technology could be employed for predicting an offshore wind turbine power converter’s lifespan. It can be performed by diagnostic health supervisory for offshore wind turbines [74].

The use of a novel simulated prototyping technique to design wind turbine blades was investigated in [75]. Numerical simulation and experimental data on turbine blade design and dynamic behavior are mixed to create an inexpensive automated digital twin model that aims to minimize uncertainty over the entire wind turbine lifespan. This model could be employed to trace and forecast changes in blade structure and evaluate its residual life.

It is difficult to foresee precisely steady structural and aerodynamic impacts, as an example of dynamic stalling and continuously increasing blade loading and thrust. It results in the unreliable design of turbines, and worst premature failures. The processes could be known by scaled testing of the wind tunnel. the interpretation of these test results is greatly enhanced by using a well-calibrated digital twin [76].

Prognostics and health management for complex and high-value equipment have been investigated in [77] using digital twin and sufficient data must be available in the wind turbine. Figure 6 represents digital twin technology in the wind turbine.

Fig. 6

Digital twin technology in the wind turbine.

In [78], a new hardware-in-loop (HIL) and software-in-loop (SIL) control, on the basis of DT technology is used. The DT concept of mentioned article intends to decrease the differentiation of the physical controller and SIL controller counterpart.

5 Ship power system

Transportation by ship is accountable for approximately 90 percent of trading all over the world, contributing to vital environmental effects. Consequently, the production of technology capable of rising ship productivity by reducing fuel consumption and excessive repair operations is a critical concern for the maritime industry [79 –82].

The ship power system has less capacity for inertia and production rather than big power systems, which normally behaves such as a microgrid. Moreover, dynamic and pulse harvest the maximum capacity in ship systems, causing serious oscillations for voltage and frequency of power system [83]. Marine surface electrification vessels reduce emissions of CO2, sulfur oxides, and nitrogen oxide. Smaller operating costs and fuel consumption could be obtained while comparing to traditional diesel-based propulsion systems. The battery control system eliminates the intermittency and provides the marine and offshore industries with dispatchable and renewable electricity. Ferryboat has the highest amount of battery-powered vessels in operation and also under construction, as it is normally used for short-distance traveling in which spends lengthier periods while charging of batteries could occur [83 –85]. In comparison, lengthier distances need bigger battery packs and demand higher prices, and at the moment, they are not desirable. During the peak period, the batteries can be utilized for handling the highest demand to control the engine load from dropping. The batteries could be employed while traveling or at the port to fuel the following systems:

sensors systems for automation operation.

power management system

marine computer system

electronic converters in the propulsion system

cooling system

fire alarm system

electric instant water-heating system

LED lighting system

solar and renewable wind system

pump system

HVAC system

the waste heat recovery system

Figure 7 represents the shipboard structure consisting of renewable energies and an energy storage system. Nowadays, the design of the shipboard power system process faces important alterations, because of adjustments in battery technology and power electronics and also using renewable energies. An emission and energy management system (EEMS) must be adopted elaborately which is a high-level control system to generate the rules for the functioning of power plants for reducing pollution and using of energy, meanwhile retaining specifications for protection and stability and executing the vessel’s mission purposes. Therefore, EEMS improves the efficiency of the propulsion system and control system by spreading the load power needed by various energy resources where optimally minimizes overall emissions and utilizes all sources. The EEMS controls fuel consumption, and emissions, system efficiency, wearing on system components, by managing the hybrid power system’s dynamic behavior. Hence it is vital to choose and develop an appropriate EEMS on the basis of the requirements of a hybrid propulsion system and objective function.

Fig. 7

Shipboard structure consisting of renewable energies and energy storage system.

The EEMS could be separated into a few groups which are shown in Fig. 8): 1) learning-based, 2) rule-based, and 3) optimization-based. The first approach is divided into dual types: stochastic and deterministic. This approach performs on several predetermined instructions. Firstly, power plants are created as specified by physical rules, also it is believed that unknown parameters are owned by such bordered sets. Then, algorithms using in deterministic control is planned to assure the plant’s steady closed-loop manner and strength to changes of parameters. However, stochastic rule-based procedures surmise, parameters and states are random parameters and matching probability density functions. Optimization-based techniques, like or dynamic programming and genetic algorithms, are usually applied in the optimization form of iterative computational. Strategies based on optimization are primarily divided into duo types: online and offline optimization. In addition, offline optimization is on the basis of predicted sailing profile data. The actual profile of sailing because of a few stochastic variables in waterways, like unforeseeable conditions of weather, marine biology elements, and random ocean currents, is very difficult to predict. An instant cost function is used to optimize the power plant online throughout the expedition. It is then possible to implement various numerical optimization methods for EEMS involving nonlinear, quadratic, linear programming with the strategy for reduction of equivalent cost. Rule-based and optimization-based plans utilize all types of predefined rules or predicted conditions for sailing. The aforementioned approaches do not automatically conform to arbitrary machine situations. So results can be yielded in short expectations. Experts have also been enthusiastic about learning-based approaches, that could be divided into a few types: unsupervised learning, reinforcement learning, and supervised learning Supervised learning algorithm is acquired by the training collection of designated instances that a qualified professional instructor offers. Unsupervised learning is learning through the exploration of the secret meaning of unlabeled data sets. These are very effective forms of learning but they are not sufficient alone to learn of the world towards real-time experiences. Also, the reinforcement learning algorithm is an aim-directed procedure in which a factor learns through interaction with the uncertain environment in real-time.

Fig. 8

The diagram showing the categorization of mainl strategies, model predictive control (MPC), equivalent cost minimization strategy(ECMS), Potryagin’s minimum principle (PMP), linear programming/quadratic programming (LP/QP).

The digital twin idea is attracting attention in the shipping industry, but it raises new concerns about data control and governance with it. It is not clear, given the wide variety of different data that can be contained in the digital twin, that all such data could be contained in the same database or in the same model of governance [86].

Industry 4.0 is now known as the fourth technological revolution in the land-based manufacturing sector following mechanized technology, mass processing, and computerization. As the maritime industry has historically adjusted and used the effects of the various technological developments, what we might call “Maritime 4.0” is now joining, which is shown in Fig. 9 [87].

Fig. 9

Shipping 4.0 [87].

A digital twin is being applied massively to different types of content as the commercial and organizational advantages of the digital twin become clear. A few instances that can be used in the knowledge are:

Designing control system and construction drawings

2-Analyzing hydrodynamic and strength data and measuring tank test

3-results of data performance in the sea trial

4-measuring data like fuel consumption, speed, actual weather from equipment

Nevertheless, it is not clear that all these various kinds of data will be contained in the same physical archive or needed to keep it. Industrial data space was developed to provide a solution to these issues. It is built on open standards and a shared governance model to implement data ownership in a distributed environment while making it easy for any trusted party to get information [88].

Through reality, ships are sailing villages with all the public facilities that we will usually find in a community, such as energy delivery, heat, and ventilation, water, and sanitation, local and external connectivity, etc. It also has systems for regulating and tracking energy generation, propulsion and steering, navigation, fire alarms, and other safety devices, and so on. Usually, these devices are not manufactured by the same vendor and this causes difficulties with the synchronization of data sources as well as control of the collected data [89 –95].

Ships have an estimated lifetime of about 30 years, and computer equipment will be upgraded several times during this life cycle. This causes extra device interfacing and interconnectivity issues. Some approaches arise from international organizations of standards, such as navigation [96] and computer data [97]. Figure 10 is an example of a ship power system using digital twin technology [98].

Fig. 10

Ship power system using digital twin technology [98].

The digital twin helps the company imagine a product’s state and condition that could be thousands of miles away. Industries exploit a digital twin in the maritime business. Big and complex boats sail across the globe and are highly efficient in operations. Industry 4.0 for manufacturing and digital twin for monitoring could be used for energy saving and better ship design [99]. The digital twin is becoming a greater priority in shipping, allowed by IoT and machine learning [100].

In [101], for a class of strict-feedback nonlinear systems with external disturbances, an event triggered robust fuzzy adaptive prescribed output finite-time control technique is introduced. To reduce the coordination burden, the relative-threshold-based event-triggered signal is implemented, and the dynamic surface control strategy is extended to solve the issue of computational complexity. For a class of strict-feedback nonlinear systems with non-affine nonlinear faults, [102] deals with investigating the problem of fuzzy adaptive control. By following the dynamic surface control technique, computational complexity is minimized. A novel fault-tolerant control technique is built within the paradigm of finite-time stability such that the closed-loop system is essentially semi-globally end-time stable, and the monitoring error converges in a finite-time to a small residual range.

6 Solar panel system

The Photovoltaic (PV) system has become a growingly common and economically viable source of electricity. In 2017 more than 100 GW of solar power plants worldwide were placed into service. Around the same time, keeping energy supply, detection, and prevention highly effective was an increasingly important issue [103 –106].

Around two percent of the solar panels are expected to fail after 12 years. At the same time, the losses incurred by deposition of dust (soiling), due to cell depletion can be an order of magnitude greater than damage [107, 108]. Due to their complex outdoor installations, an increased number of power electronic converters at the PV panel level, tough mission profiles, manufacturing defects, and aging are susceptible to a variety of faults that occurred in the PV system. In the PV system, Table 1 represents common fault modes for each component.

Table 1
Standard fault modes of a PV system

Components Common fault modes

Sensors [109] Open circuit

Wear-out (adjusting parameter drift)

Electrical converters [110 –112] Switch open circuit

Switch short circuit

Capacitor or inductor deterioration

PV panel [113 –116] Making dirty

partial shading conditions

Cells breakage

EVA browning

Bypass diode faults

Potential induced degradation (PID)

Interconnect failures

Internal faults corresponding to PV shunt and series resistances

Components	Common fault modes
Sensors [109]	Open circuit
	Wear-out (adjusting parameter drift)
Electrical converters [110 –112]	Switch open circuit
	Switch short circuit
	Capacitor or inductor deterioration
PV panel [113 –116]	Making dirty
	partial shading conditions
	Cells breakage
	EVA browning
	Bypass diode faults
	Potential induced degradation (PID)
	Interconnect failures
	Internal faults corresponding to PV shunt and series resistances

Thus, it must not be underestimated the value of deciding the appropriate cleaning action. Rising vegetation may also be causing shading losses. These and several other issues can be prevented or can considerably by early identification and effective maintenance actions. Installing a data acquisition system that tracks different parameters of Photovoltaic (PV) equipment is a popular method for keeping track of solar plant safety. Such research is usually performed top-down rather than bottom-up, which will be more useful because it is the components that are more vulnerable to error.

The fundamental monitoring system helps each solar panel to gather data and track [117, 118], as well as make this data accessible [119] for analysis and prediction.

Big data analytics engine is another significant part of the program, which is shown in Fig. 12. It performs classification tasks using model variables values and offers useful machine safety metrics. Adaptive neuro-fuzzy inference system (ANFIS) and genetic algorithm (GA) can be used to estimate optimal photovoltaic system parameters in grid-connected mode, and the inverter adjusts the output power of the system [120, 121].

Fig. 11

Big data analytics engine.

Fig. 12

Equivalent Circuit of Solar Cell system.

In fact, the solar cell is a wide area p-n diode with the junction located close to the top surface. So a current source model of ideal solar cell parallel with a diode that mathematically represents the characteristic of I-V by: $I = I_{{PV}_{cell}} - I_{d} = I_{{PV}_{cell}} - I_{0_{cell}}$ (6)

In which I_PVcell is the current produced by the incident light and I_d is equal to Shockley diode equation and I_0cell is the reverse saturation or leakage current of the diode and q is equal to the electron charge (1.60217646×10^–19C), k is the Boltzmann constant (1.3806503×10^–23J/K), T(K) is the temperature of the p-n junction a is equal to the diode ideality constant. Also, the model is supplemented with a shunt and a series resistance component, as no solar cell is ideal in operation. Figure 12 presents the equivalent Circuit of the Solar Cell system.

All components at PV Digital twin network work together to turn endless data tracking flow into series of model variables that are then converted into network health metrics and finally into human-readable information to make successful business decisions. The system’s block scheme as a whole is given in Fig. 13.

Fig. 13

Digital twin data flow in solar panel system [122].

In [122], a digital twin is developed that measures the observable characteristic performance of a PV energy conversion unit in real-time, which is shown in Fig. 14. The diagnosis of fault is achieved by creating and measuring a residual error matrix, which is the difference between the expected and calculated outputs. Finally, researchers in [123 –125] explore different methods for the fault analysis of electronic power converters in PV systems. Likewise, researchers in [126 –128] present specific methodologies for the evaluation of PV panel faults.

Fig. 14

A description of a digital twin approach to the full PV method for fault diagnosis [122].

7 DC-DC power converter system

The power semiconductor and capacitor account for over half of failure, according to an industry-based survey [129 –133]. So, tracking the health status of these two elements is important for preventing unintended faults and contributing to high maintenance costs. Measurement circuit-based condition control approaches for assessing the safety of power semiconductor switches and capacitors, respectively, have been suggested [134 –136]. For power semiconductor, like on-state resistance/voltage, threshold voltage, and miller plateau in turn-on gate voltage and the equivalent series resistance (ESR) and capacitance for capacitor has been used [137 –141]. These approaches are useful for tracking power semiconductors and capacitors’ degradation. However, additional hardware circuits are required for each part, respectively. In addition, the added circuit failure can cause a device to fail to be controlled. Model-based parameter recognition is a promising tool for condition tracking to resolve the above challenges. Then, some algorithms measure the coefficients of the transfer function, such as recursive least square (RLS) in [142], and Kalman Filter (KF) in [143, 144]. This approach is successful in optimizing system efficiency, but it is not ideal for tracking the condition. In [145], a simple boost converter model is developed and the inductance and capacitance are determined using a generalized gradient descent algorithm. A buck converter model is investigated in [146], using the Biogeography-based Optimization (BBO) approach to define the internal parameters. The biggest problem with the above approaches is that they need to reach the controller to get the service cycle ratio (modulation signal) in real-time. Since only the converter model is installed in the above methods while the configuration and controller unit of the sampling circuit is not used. Furthermore, the above approaches will insert additional signals into the controller which is intrusive to the device.

The digital twin can be used for the supervisory of a DC-DC power converter to resolve the aforementioned problems. A buck converter is one example of analyzing the digital twin performance. The controller and sampling circuit and power stage are the parts of the system that the digital twin is executed. The capacitor and MOSFET damage the circuit, and they need to monitor the health circumstance. Moreover, Particle swarm optimization is implemented to upgrade the value of internal parameters of digital twin by measuring from physical DC-DC converter which is shown in Fig. 15. It has the ability to recognize the degradation of components by data cluster-based method without using additional circuits [147].

Fig. 15

The structure of the digital twin that is implemented in the buck converter [147].

The digital twin can be used for diagnostic of power electronic converter and it can act as a real-time probabilistic. The structure of the digital twin in [148] separates into subsystem control layers and using generalized polynomial chaos expansion in order to have a better performance.

8 Disadvantages of digital twin

Besides the entire advantages of the digital twin, it is vital to express the disadvantages in order to have a better assessment. Four challenges would be examined, which are technical, workforce, business, and social.

Two subjects are relevant to technical challenges. Model composability is the first issue that must be dealing with. Because the magnitude and complexity of combining models have increased and it is essential to have an interoperability standard. The second issue is that digitalizing legacy products are resource-intensive and difficult to implement due to a lack of information.

The next problem is related to workforce challenges. Digitalizing legacy products puts extra demand on the experts involved in creating them, and lack of knowledge and skills in the current workforce for digital engineering makes it more difficult. It is considered that digital experts might be least experienced in the business domain and products. Therefore, the high demand for talent makes hiring difficult. Another problem is that organizational structures typically are not suitable for creating the digital thread.

Business challenges are related to two issues. First, changing the traditional business model which is used for a long time, is hard work, and besides, inversion of the value of product and model might have happened. The final problem is related to social challenges. Sometimes, it is excruciating to get people to give up their models and data, and the willingness to accept models as the primary source of truth takes so much time.

9 Conclusion

Due to the growth of the industries in recent years, system evaluation is necessary to diagnose and detect problems in the power system. Digital twin (DT) not only can be used to simulate the performance of the system but also gives insights for better decision making at every level of system operation. Obviously, implementing advanced analytic methodologies offer higher accuracy than the conventional power system without imposing the cost of adding or replacing existing sensors. The DT platform represents the essence of the modern power system by integrating domain knowledge with big data analytics and a large-scale network system that delivers value from multiple aspects for experts. Particularly, in this paper, the applications of digital twin are discussed entirely in renewable energies, shipboard, and power electronic systems. Moreover, the potential benefits and drawbacks of these technologies also have been entirely expressed from various aspects. Finally, it can be understood that using digital twins in various types of systems leads to better efficiency and detection of parameter faults and diagnosis.

References

Rosen

, Von Wichert

, Lo

and Bettenhausen

K.D.

, About the importance of autonomy and digital twins for the future of manufacturing, IFAC-PapersOnLine 48 (2015), 567–572.

Parvaresh

, Abrazeh

, Mohseni

S.-R.

, Zeitouni

M.J.

, Gheisarnejad

and Khooban

M.-H.

, A Novel Deep Learning Backstepping Controller-Based Digital Twins Technology for Pitch Angle Control of Variable Speed Wind Turbine, Designs 4 (2020), 15.

Glaessgen

and Stargel

, The digital twin paradigm for future NASA and US Air Force vehicles, in 53rd AIAA/ASME/ASCE/AHS/ASC structures, structural dynamics and materials conference 20th AIAA/ASME/AHS adaptive structures conference 14th AIAA, 2012, p. 1818.

Tao

, Cheng

, Qi

, Zhang

and Sui

, Digital twin-driven product design, manufacturing and service with big data, The International Journal of Advanced Manufacturing Technology 94 (2018), 3563–3576.

Söderberg

, Wärmefjord

, Carlson

J.S.

and Lindkvist

, Toward a Digital Twin for real-time geometry assurance in individualized production, CIRP Annals 66 (2017), 137–140.

and Xu

, Cloud-based manufacturing equipment and big data analytics to enable on-demand manufacturing services, Robotics and Computer-Integrated Manufacturing 57 (2019), 92–102.

Wells

L.J.

, Camelio

J.A.

, Williams

C.B.

and White

, Cyber-physical security challenges in manufacturing systems, Manufacturing Letters 2 (2014), 74–77.

Ashton

, That ‘internet of things’ thing, RFID Journal 22 (2009), 97–114.

Sarma

, Brock

D.L.

and Ashton

, The networked physical world, Auto-ID Center White Paper MIT-AUTOID-WH-001, 2000.

10.

Jazdi

, Cyber physical systems in the context of Industry 4.0, in 2014 IEEE International Conference on Automation, Quality and Testing, Robotics, 2014, pp. 1–4.

11.

Rainer

and Alexander

, Industrie 4.0: hit or hype? Industrial Electronics Magazine 8 (2014), 56–58.

12.

Gheisarnejad

and Khooban

M.H.

, IoT-Based DC/DC Deep Learning Power Converter Control: Real-Time Implementation, IEEE Transactions on Power Electronics 35 (2020), 13621–13630.

13.

Wang

, Yang

, Wang

, Sherratt

R.S.

and Zhang

, Big data service architecture: a survey, Journal of Internet Technology 21 (2020), 393–405.

14.

Grieves

and Vickers

, Digital twin: Mitigating unpredictable, undesirable emergent behavior in complex systems, in Transdisciplinary Perspectives on Complex Systems, ed: Springer, 2017, pp. 85–113.

15.

Wenger

, Artificial intelligence and tutoring systems: computational and cognitive approaches to the communication of knowledge, Morgan Kaufmann, 2014.

16.

Bauer

, Wollherr

and Buss

, Human–robot collaboration: a survey, International Journal of Humanoid Robotics 5 (2008), 47–66.

17.

, Liu

, Kevin

, Wang

, Huang

and Xu

, Digital Twin-driven smart manufacturing: Connotation, reference model, applications and research issues, Robotics and Computer-Integrated Manufacturing 61 (2020), 101837.

18.

, Peng

and Xu

, Energy-efficient cyber-physical production network: Architecture and technologies, Computers & Industrial Engineering 129 (2019), 56–66.

19.

Kumar

, From mass customization to mass personalization: a strategic transformation, International Journal of Flexible Manufacturing Systems 19 (2007), 533–547.

20.

Gheisarnejad

, Mohammadi-Moghadam

, Boudjadar

and Khooban

M.H.

, Active power sharing and frequency recovery control in an islanded microgrid with nonlinear load and nondispatchable DG, IEEE Systems Journal 14 (2019), 1058–1068.

21.

Gheisarnejad

, Karimaghaee

, Boudjadar

and Khooban

M.-H.

, Real-time cellular wireless sensor testbed for frequency regulation in smart grids, IEEE Sensors Journal 19 (2019), 11656–11665.

22.

Khooban

M.H.

and Gheisarnejad

, Islanded microgrid frequency regulations concerning the integration of tidal power units: real-time implementation, IEEE Transactions on Circuits and Systems II: Express Briefs 67 (2019), 1099–1103.

23.

Gheisarnejad

and Khooban

M.H.

, Secondary load frequency control for multi-microgrids: HiL real-time simulation, Soft Computing 23 (2019), 5785–5798.

24.

, Ai

, Qiu

R.C.

and Zhang

, Preliminary Exploration on Digital Twin for Power Systems: Challenges, Framework, and Applications, arXiv preprint arXiv:1909.06977, 2019.

25.

Xiu

and Karniadakis

G.E.

, The Wiener–Askey polynomial chaos for stochastic differential equations, SIAM Journal on Scientific Computing 24 (2002), 619–644.

26.

Benigni

and Monti

, A parallel approach to real-time simulation of power electronics systems, IEEE Transactions on Power Electronics 30 (2014), 5192–5206.

27.

Milton

, Benigni

and Bakos

, System-level, FPGA-based, real-time simulation of ship power systems, IEEE Transactions on Energy Conversion 32 (2017), 737–747.

28.

Difronzo

, Milton

, Davidson

and Benigni

, Hardware-in-the-loop testing of high switching frequency power electronics converters, in 2017 IEEE Electric Ship Technologies Symposium (ESTS), 2017, pp. 299–304.

29.

Milton

, Benigni

and Monti

, Real-time multi-FPGA simulation of energy conversion systems, IEEE Transactions on Energy Conversion 34 (2019), 2198–2208.

30.

Milton

, Benigni

, Vygoder

, Gudex

and Cuzner

, Power electronic system real-time simulation on national instruments FPGA platforms, in 2019 IEEE Electric Ship Technologies Symposium (ESTS), 2019, pp. 32–38.

31.

Abrazeh

, Parvaresh

, Mohseni

S.-R.

, Zeitouni

M.J.

, Gheisarnejad

and Khooban

M.H.

, Nonsingular Terminal Sliding Mode Control With Ultra-Local Model and Single Input Interval Type-2 Fuzzy Logic Control for Pitch Control of Wind Turbines, IEEE/CAA Journal of Automatica Sinica 8 (2021), 690–700.

32.

Dudley

, BP statistical review of world energy 2016, London, UK, 2015.

33.

Parobek

, Paluš

, Kalamárová

, Loučanová

, Šupín

, Križanová

, et al., Energy Utilization of Renewable Resources in the European Union–Cluster Analysis Approach, BioResources 11 (2016), 984–995.

34.

Ghosh

, Isbeih

Y.J.

, Bhattarai

, El Moursi

M.S.

, El-Saadany

E.F.

and Kamalasadan

, A dynamic coordination control architecture for reactive power capability enhancement of the DFIG-based wind power generation, IEEE Transactions on Power Systems 35 (2020), 3051–3064.

35.

Hansen

M.O.

, Aerodynamics of wind turbines: Routledge, 2015.

36.

A.M. Update, Global wind report, Global Wind Energy Council, 2017.

37.

Rezaei

, Advanced control of wind turbines: Brief survey, categorization, and challenges, in 2015 American Control Conference (ACC), 2015, pp. 3044–3051.

38.

Gharibeh

H.F.

, Khiavi

L.M.

, Farrokhifar

, Alahyari

and Pozo

, Capacity value of variable-speed wind turbines, in 2019 IEEE Milan PowerTech, 2019, pp. 1–5.

39.

Chen

, Chen

and Gong

, New overall power control strategy for variable-speed fixed-pitch wind turbines within the whole wind velocity range, IEEE Transactions on Industrial Electronics 60 (2012), 2652–2660.

40.

Miao

, Wen

, Xie

, Yue

and Lee

W.-J.

, Coordinated control strategy of wind turbine generator and energy storage equipment for frequency support, IEEE Transactions on Industry Applications 51 (2015), 2732–2742.

41.

Xie

, Jiang

and Li

, Effect of wind speed on wind turbine power converter reliability, IEEE Transactions on Energy Conversion 27 (2012), 96–104.

42.

Anjun

, Hao

, Shuju

and Honghua

, Pitch control of large scale wind turbine based on expert PID control, in 2011 International Conference on Electronics, Communications and Control (ICECC), 2011, pp. 3836–3839.

43.

Kim

J.-S.

, Jeon

and Heo

, Design of adaptive PID for pitch control of large wind turbine generator, in 2011 10th International Conference on Environment and Electrical Engineering, 2011, pp. 1–4.

44.

Jeon

and Paek

, Design and Verification of the LQR Controller Based on Fuzzy Logic for Large Wind Turbine, Energies 14 (2021), 230.

45.

Moghadam

H.M.

, Gheisarnejad

, Esfahani

and Khooban

M.-H.

, A Novel Supervised Control Strategy for Interconnected DFIG-Based Wind Turbine Systems: MiL Validations, IEEE Transactions on Emerging Topics in Computational Intelligence, 2020.

46.

Bakka

, Karimi

H.-R.

and Christiansen

, Linear parameter-varying modelling and control of an offshore wind turbine with constrained information, IET Control Theory & Applications 8 (2014), 22–29.

47.

Boukhezzar

and Siguerdidjane

, Nonlinear control of a variable-speed wind turbine using a two-mass model, IEEE Transactions on Energy Conversion 26 (2010), 149–162.

48.

, Yan

, Shaltout

M.L.

and Chen

, Optimal real-time control of wind turbine during partial load operation, IEEE Transactions on Control Systems Technology 23 (2015), 2216–2226.

49.

Da Costa

J.P.

, Pinheiro

, Degner

and Arnold

, Robust controller for DFIGs of grid-connected wind turbines, IEEE Transactions on Industrial Electronics 58 (2010), 4023–4038.

50.

Beltran

, Ahmed-Ali

and Benbouzid

M.E.H.

, Sliding mode power control of variable-speed wind energy conversion systems, IEEE Transactions on Energy Conversion 23 (2008), 551–558.

51.

Dang

, Wang

and Cai

, Offset-free predictive control for variable speed wind turbines, IEEE Transactions on Sustainable Energy 4 (2012), 2–10.

52.

Civelek

, Lüy

, Çam

and Barışçı

, Control of pitch angle of wind turbine by fuzzy PID controller, Intelligent Automation & Soft Computing 22 (2016), 463–471.

53.

Simão

, Powell

, Archer

and Kempton

, The challenge of integrating offshore wind power in the US electric grid. Part II: Simulation of electricity market operations, Renewable Energy 103 (2017), 418–431.

54.

Sahnoun

M.H.

, Baudry

, Mustafee

, Louis

, Smart

P.A.

, Godsiff

, et al., Modelling and simulation of operation and maintenance strategy for offshore wind farms based on multi-agent system, Journal of Intelligent Manufacturing 30 (2019), 2981–2997.

55.

Dulman

M.T.

and Gupta

S.M.

, Maintenance and remanufacturing strategy: using sensors to predict the status of wind turbines, Journal of Remanufacturing 8 (2018), 131–152.

56.

Dong

, Time-domain fatigue response and reliability analysis of offshore wind turbines with emphasis on welded tubular joints and gear components, 2012.

57.

Zhang

, Chen

, Matveev

, Nilssen

and Nysveen

, High-power generators for offshore wind turbines, Energy Procedia 35 (2013), 52–61.

58.

Dao

, Kazemtabrizi

and Crabtree

, Wind turbine reliability data review and impacts on levelised cost of energy, Wind Energy 22 (2019), 1848–1871.

59.

Tavner

, Faulstich

, Hahn

and van Bussel

, Reliability & availability of wind turbine electrical & electronic components, EPE Journal 20 (2010), 45–50.

60.

Shoaib

, Siddiqui

, Rehman

, Khan

and Alhems

L.M.

, Assessment of wind energy potential using wind energy conversion system, Journal of Cleaner Production 216 (2019), 346–360.

61.

Ribrant

and Bertling

, Survey of failures inwind power systems with focus on Swedish wind power plants during 1997–2005, in 2007 IEEE Power Engineering Society General Meeting, 2007, pp. 1–8.

62.

Velarde

, Kramhøft

, Sørensen

J.D.

and Zorzi

, Fatigue reliability of large monopiles for offshore wind turbines, International Journal of Fatigue 134 (2020), 105487.

63.

, Gear failure modes—importance of lubrication and mechanics, ASLe Transactions 19 (1976), 239–249.

64.

Rademakers

L.L.

, Braam

H.H.

and Obdam

T.T.

, Operation and maintenance of offshore wind energy systems, in Wind Energy Systems, ed: Elsevier, 2011, pp. 546–583.

65.

Márquez

F.P.G.

, Tobias

A.M.

, Pérez

J.M.P.

and Papaelias

, Condition monitoring of wind turbines: Techniques and methods, Renewable Energy 46 (2012), 169–178.

66.

Nejad

A.R.

, Gao

and Moan

, Fatigue reliability-based inspection and maintenance planning of gearbox components in wind turbine drivetrains, Energy Procedia 53 (2014), 248–257.

67.

Nejad

A.R.

and Moan

, On model-based system approach for health monitoring of drivetrains in floating wind turbines, Procedia Engineering 199 (2017), 2202–2207.

68.

Straiton

, Neural–digital interfaces: creating bionic humans, ed: Future Science, 2020.

69.

Kamath

, Morgan

and Ali

M.I.

, Industrial IoT and Digital Twins for a Smart Factory: An open source toolkit for application design and benchmarking, in 2020 Global Internet of Things Summit (GIoTS), 2020, pp. 1–6.

70.

Roy

, Stark

, Tracht

, Takata

and Mori

, Continuous maintenance and the future–Foundations and technological challenges, Cirp Annals 65 (2016), 667–688.

71.

Loving

, Crofts

, Sykes

, Iglesias

, Coleman

, Thomas

, et al., Pre-conceptual design assessment of DEMO remote maintenance, Fusion Engineering and Design 89 (2014), 2246–2250.

72.

Etxegarai

, Valverde

, Eguia

and Perea

, Analysis of loss of life of dry-type WTSU transformers in offshore wind farms, International Journal of Renewable Energy Research, 2020.

73.

Sivalingam

, Sepulveda

, Spring

and Davies

, A review and methodology development for remaining useful life prediction of offshore fixed and floating wind turbine power converter with digital twin technology perspective, in 2018 2nd International Conference on Green Energy and Applications (ICGEA), 2018, pp. 197–204.

74.

Baldassarre

, Ceruti

, Valyou

D.N.

and Marzocca

, Towards a digital twin realization of the blade system design study wind turbine blade, Wind and Structures 28 (2019), 271–284.

75.

LeBlanc

and Ferreira

, Experimental characterization of H-VAWT turbine for development of a digital twin, in Journal of Physics: Conference Series, 2020, p. 012057.

76.

Tao

, Zhang

, Liu

and Nee

, Digital twin driven prognostics and health management for complex equipment, Cirp Annals 67 (2018), 169–172.

77.

Meraghni

, Terrissa

L.S.

, Yue

, Ma

, Jemei

and Zerhouni

, A data-driven digital-twin prognostics method for proton exchange membrane fuel cell remaining useful life prediction, International Journal of Hydrogen Energy 46 (2021), 2555–2564.

78.

Zeitouni

M.J.

, Parvaresh

, Abrazeh

, Mohseni

S.-R.

, Gheisarnejad

and Khooban

M.-H.

, Digital twins-assisted design of next-generation advanced controllers for power systems and electronics: Wind turbine as a case study, Inventions 5 (2020), 19.

79.

Coraddu

, Oneto

, Baldi

, Cipollini

, Atlar

and Savio

, Data-driven ship digital twin for estimating the speed loss caused by the marine fouling, Ocean Engineering 186 (2019), 106063.

80.

Hasanvand

, Rafiei

, Gheisarnejad

and Khooban

M.-H.

, Reliable Power Scheduling of an Emission-Free Ship: Multiobjective Deep Reinforcement Learning, IEEE Transactions on Transportation Electrification 6 (2020), 832–843.

81.

Gheisarnejad

, Khooban

M.H.

and Boudjadar

, Adaptive network based fuzzy inference system for frequency regulation in modern maritime power systems, in 2019 IEEE5th International forumonResearch and Technology for Society and Industry (RTSI), 2019, pp. 302–307.

82.

Mosayebi

, Sadeghzadeh

S.M.

, Gheisarnejad

and Khooban

M.H.

, Intelligent and Fast Model-Free Sliding Mode Control for Shipboard DC Microgrids, IEEE Transactions on Transportation Electrification, 2020.

83.

Moghadam

H.M.

, Khooban

M.H.

, Dragicevic

and Masoudian

, Using Interval Type2 FuzzyController in Ship Power Systems in Presence of Pulsed Power Loads, in 2018 IEEE International Conference on Electrical Systems for Aircraft, Railway, Ship Propulsion and Road Vehicles & International Transportation Electrification Conference (ESARS-ITEC), 2018, pp. 1–6.

84.

Vafamand

, Mardani

M.M.

, Khooban

M.H.

, Blaabjerg

and Boudjadar

, Pulsed power load effect mitigation in DC shipboard microgrids: a constrained model predictive approach, IET Power Electronics 12 (2019), 2155–2160.

85.

Yousefizadeh

, Bendtsen

J.D.

, Vafamand

, Khooban

M.H.

, Dragičević

and Blaabjerg

, EKF-based predictive stabilization of shipboard DC microgrids with uncertain time-varying load, IEEE Journal of Emerging and Selected Topics in Power Electronics 7 (2018), 901–909.

86.

Aiello

, Giallanza

and Mascarella

, Towards Shipping 4.0. A preliminary gap analysis, Procedia Manufacturing 42 (2020), 24–29.

87.

Soldi

, Gaglione

, Forti

, Di Simone

, Daffinà

F.C.

, Bottini

, et al., Space-based Global Maritime Surveillance. Part I: Satellite Technologies, arXiv preprint arXiv:2011.11304, 2020.

88.

Hou

, Song

, Hofmann

H.F.

and Sun

, Control strategy for battery/flywheel hybrid energy storage in electric shipboard microgrids, IEEE Transactions on Industrial Informatics 17 (2020), 1089–1099.

89.

Chen

C.-J.

, Su

C.-L.

and Teng

J.-H.

, Electrical Load Analysis for Shipboard Power Systems Using Load Survey Data, IEEE Transactions on Industry Applications 56 (2020), 1180–1189.

90.

Soliman

, Barone

and Frangopol

D.M.

, Fatigue reliability and service life prediction of aluminum naval ship details based on monitoring data, Structural Health Monitoring 14 (2015), 3–19.

91.

Khooban

M.-H.

, Dragicevic

, Blaabjerg

and Delimar

, Shipboard microgrids: A novel approach to load frequency control, IEEE Transactions on Sustainable Energy 9 (2017), 843–852.

92.

Hammond

, Lacey

, Gilles

, Viquerat

, Börjesson

, Herr

, et al., Estimates of cetacean abundance in European Atlantic waters in summer 2016 from the SCANS-III aerial and shipboard surveys, Wageningen Marine Research, 2017.

93.

Gheisarnejad

, Khooban

M.-H.

, Dragičević

and Boudjadar

, Shipboard Secondary Load Frequency Control Based on PPLs and Communication Degradations, in IECON 2019-45th Annual Conference of the IEEE Industrial Electronics Society, 2019, pp. 4063–4067.

94.

Khooban

M.H.

, Gheisarnejad

, Farsizadeh

, Masoudian

and Boudjadar

, A new intelligent hybrid control approach for DC–DC converters in zero-emission ferry ships, IEEE Transactions on Power Electronics 35 (2019), 5832–5841.

95.

Heydari

, Gheisarnejad

, Khooban

M.H.

, Dragicevic

and Blaabjerg

, Robust and fast voltage-source-converter (VSC) control for naval shipboard microgrids, IEEE Transactions on Power Electronics 34 (2019), 8299–8303.

96.

Discenzo

F.M.

, Loparo

K.A.

, Cassar

and Chung

, Machinery condition monitoring using wireless self-powered sensor nodes, in Proc. 24th Int. Modal Analysis Conf, 2006.

97.

Inozu

, Schaedel

, Roy

and Molinari

, Sharing ships’ reliability availability maintainability (RAM) information to improve cost effectiveness and safety, Advances in Safety and Reliability. Oxford: Pergamon, pp. 1039–1046, 1997.

98.

Tao

, Sui

, Liu

, Qi

, Zhang

, Song

, et al., Digital twin-driven product design framework, International Journal of Production Research 57 (2019), 3935–3953.

99.

Halff

, Younes

and Boersma

, The likely implications of the new IMO standards on the shipping industry, Energy Policy 126 (2019), 277–286.

100.

Madni

A.M.

, Madni

C.C.

and Lucero

S.D.

, Leveraging digital twin technology in model-based systems engineering, Systems 7 (2019), 7.

101.

Sun

, Jianbin

, Karimi

H.R.

and Fu

, Event-triggered robust fuzzy adaptive finite-time control of nonlinear systems with prescribed performance, IEEE Transactions on Fuzzy Systems, 2020.

102.

Sun

, Liu

, Qiu

and Feng

, Fuzzy adaptive finite-time fault-tolerant control for strict-feedback nonlinear systems, IEEE Transactions on Fuzzy Systems, 2020.

103.

Wang

, Wang

, Yang

, Zhu

and Liu

, Big data driven Hierarchical Digital Twin Predictive Remanufacturing paradigm: Architecture, control mechanism, application scenario and benefits, Journal of Cleaner Production 248 (2020), 119299.

104.

López-Erauskin

, González

, Petrone

, Spagnuolo

and Gyselinck

, Multi-Variable Perturb and Observe Algorithm for Grid-Tied PV Systems With Joint Central and Distributed MPPT Configuration, IEEE Transactions on Sustainable Energy 12 (2020), 360–367.

105.

Rakhshan

, Vafamand

, Khooban

M.-H.

and Blaabjerg

, Maximum power point tracking control of photovoltaic systems: A polynomial fuzzy model-based approach, IEEE Journal of Emerging and Selected Topics in Power Electronics 6 (2017), 292–299.

106.

Heydari-doostabad

, Keypour

, Khalghani

M.R.

and Khooban

M.H.

, A new approach in MPPT for photovoltaic array based on extremum seeking control under uniform and non-uniform irradiances, Solar Energy 94 (2013), 28–36.

107.

Reinders

, Moser

, Van Sark

, Oreski

, Pearsall

, Scognamiglio

, et al., Introducing ‘PEARL-PV’: Performance and Reliability of Photo-voltaic Systems: Evaluations of Large-Scale Monitoring Data, in 2018 IEEE 7thWorld Conference on Photovoltaic Energy Conversion (WCPEC)(A Joint Conference of 45th IEEE PVSC, 28th PVSEC & 34th EU PVSEC), 2018, pp. 0762–0766.

108.

Mejia

, Kleissl

and Bosch

, The effect of dust on solar photovoltaic systems, Energy Procedia 49 (2014), 2370–2376.

109.

Poon

, Jain

, Konstantakopoulos

I.C.

, Spanos

, Panda

S.K.

and Sanders

S.R.

, Model-based fault detection and identification for switching power converters, IEEE Transactions on Power Electronics 32 (2016), 1419–1430.

110.

Wang

, Liserre

and Blaabjerg

, Toward reliable power electronics: Challenges, design tools, and opportunities, IEEE Industrial Electronics Magazine 7 (2013), 17–26.

111.

Wang

, Liserre

, Blaabjerg

, de Place Rimmen

, Jacobsen

J.B.

, Kvisgaard

, et al., Transitioning to physics-of-failure as a reliability driver in power electronics, IEEE Journal of Emerging and Selected Topics in Power Electronics 2 (2013), 97–114.

112.

Yang

, Xiang

, Bryant

, Mawby

, Ran

and Tavner

, Condition monitoring for device reliability in power electronic converters: A review, IEEE Transactions on Power Electronics 25 (2010), 2734–2752.

113.

Morlier

, Siebert

, Kunze

, Mathiak

and Köntges

, Detecting photovoltaic module failures in the field during daytime with ultraviolet fluorescence module inspection, IEEE Journal of Photovoltaics 7 (2017), 1710–1716.

114.

Alam

M.K.

, Khan

F.H.

, Johnson

and Flicker

, PV faults:Overview,modeling, prevention and detection techniques, in 2013 IEEE 14th Workshop on Control and Modeling for Power Electronics (COMPEL), 2013, pp. 1–7.

115.

Sharma

and Chandel

, Performance and degradation analysis for long term reliability of solar photovoltaic systems: A review, Renewable and Sustainable Energy Reviews 27 (2013), 753–767.

116.

Lee

, Ko

S.-K.

and Lee

B.-T.

, Forecast uncertainty-based performance degradation diagnosis of solar PV systems, IET Renewable Power Generation 14 (2020), 968–975.

117.

Caine

H.R.

, System and method for monitoring performance of a photovoltaic array, ed: Google Patents, 2014.

118.

Amin

, Yi

L.Z.

and Sopian

, Microcontroller based smart charge controller for standalone solar photovoltaic power systems, in 2009 34th IEEE Photovoltaic Specialists Conference (PVSC), 2009, pp. 001094–001097.

119.

Sarkar

, Rao

K.U.

, Bhargav

, Sheshaprasad

and CA

A.S.

, IoT Based Wireless Sensor Network (WSN) for Condition Monitoring of Low Power Rooftop PV Panels, in 2019 IEEE 4th International Conference on Condition Assessment Techniques in Electrical Systems (CATCON), 2019, pp. 1–5.

120.

Shaiek

, Smida

M.B.

, Sakly

and Mimouni

M.F.

, Comparison between conventional methods and GA approach for maximum power point tracking of shaded solar PV generators, Solar Energy 90 (2013), 107–122.

121.

Rezvani

, Moghadam

H.M.

, Khalili

and Mohammadinodoushan

, Comparison Study of Maximum Power Point Tracker Techniques for PV Systems in the Grid Connected Mode, International Journal of Review in Life Sciences 5 (2015), 1175–1184.

122.

Jain

, Poon

, Singh

J.P.

, Spanos

, Sanders

S.R.

and Panda

S.K.

, A digital twin approach for fault diagnosis in distributed photovoltaic systems, IEEE Transactions on Power Electronics 35 (2019), 940–956.

123.

Ribeiro

, Cardoso

A.J.M.

and Boccaletti

, Fault-tolerant strategy for a photovoltaic DC–DC converter, IEEE Transactions on Power Electronics 28 (2012), 3008–3018.

124.

Jamshidpour

, Poure

and Saadate

, Photovoltaic systems reliability improvement by real-time FPGA-based switch failure diagnosis and fault-tolerant DC–DC converter, IEEE Transactions on Industrial Electronics 62 (2015), 7247–7255.

125.

Rahman

M.A.

, Pazouki

, Sozer

and De Abreu-Garcia

J.A.

, Fault Detection of Switch Mode Power Converters Based on Radiated EMI Analysis, in 2019 IEEE Energy Conversion Congress and Exposition (ECCE), pp. 2968–2972.

126.

Kim

K.A.

, Seo

G.-S.

, Cho

B.-H.

and Krein

P.T.

, Photovoltaic hot-spot detection for solar panel substrings using AC parameter characterization, IEEE Transactions on Power Electronics 31 (2015), 1121–1130.

127.

Solórzano

and Egido

, Automatic fault diagnosis in PV systems with distributed MPPT, Energy Conversion and Management 76 (2013), 925–934.

128.

Wang

, Liu

A.C.-F.

, Chung

H.S.-H.

, Lau

R.W.-H.

, Zhang

and Lo

A.W.-L.

, Fault diagnosis of photovoltaic panels using dynamic current–voltage characteristics, IEEE Transactions on Power Electronics 31 (2015), 1588–1599.

129.

Yang

, Bryant

, Mawby

, Xiang

, Ran

and Tavner

, An industry-based survey of reliability in power electronic converters, IEEE Transactions on Industry Applications 47 (2011), 1441–1451.

130.

Gheisarnejad

, Farsizadeh

, Tavana

M.-R.

and Khooban

M.H.

, A Novel Deep Learning Controller for DC/DC Buck-Boost Converters inWireless Power Transfer Feeding CPLs, IEEE Transactions on Industrial Electronics, 2020.

131.

Hajihosseini

, Andalibi

, Gheisarnejad

, Farsizadeh

and Khooban

M.-H.

, DC/DC power converter control-based deep machine learning techniques: Real-time implementation, IEEE Transactions on Power Electronics 35 (2020), 9971–9977.

132.

Farsizadeh

, Gheisarnejad

, Mosayebi

, Rafiei

and Khooban

M.H.

, An intelligent and fast controller for DC/DC converter feeding CPL in a DC microgrid, Ieee Transactions on Circuits and Systems Ii: Express Briefs 67 (2019), 1104–1108.

133.

Gheisarnejad

, Farsizadeh

and Khooban

M.H.

, A novel non-linear deep reinforcement learning controller for DC/DC power buck converters, IEEE Transactions on Industrial Electronics, 2020.

134.

Wang

, Cai

, Du

and Zhou

, Review of power semiconductor device reliability for power converters, CPSS Transactions on Power Electronics and Applications 2 (2017), 101–117.

135.

Soliman

, Wang

and Blaabjerg

, A review of the condition monitoring of capacitors in power electronic converters, IEEE Transactions on Industry Applications 52 (2016), 4976–4989.

136.

Choi

U.-M.

, Jørgensen

and Blaabjerg

, Advanced accelerated power cycling test for reliability investigation of power device modules, IEEE Transactions on Power Electronics 31 (2016), 8371–8386.

137.

Dusmez

and Akin

, An accelerated thermal aging platform to monitor fault precursor on-state resistance, in 2015 IEEE International Electric Machines & Drives Conference (IEMDC), 2015, pp. 1352–1358.

138.

Astigarraga

, Ibanez

F.M.

, Galarza

, Echeverria

J.M.

, Unanue

, Baraldi

, et al., Analysis of the results of accelerated aging tests in insulated gate bipolar transistors, IEEE Transactions on Power Electronics 31 (2015), 7953–7962.

139.

, Chen

, Wang

, Zhai

and Vachtsevanos

G.J.

, Online condition monitoring of power MOSFET gate oxide degradation based on miller platform voltage, IEEE Transactions on Power Electronics 32 (2016), 4776–4784.

140.

Vogelsberger

M.A.

, Wiesinger

and Ertl

, Life-cycle monitoring and voltage-managing unit for DC-link electrolytic capacitors in PWM converters, IEEE Transactions on Power Electronics 26 (2010), 493–503.

141.

Agarwal

, Ahmad

M.W.

and Anand

, Quasi-online technique for health monitoring of capacitor in single-phase solar inverter, IEEE Transactions on Power Electronics 33 (2017), 5283–5291.

142.

Algreer

, Armstrong

and Giaouris

, Active online system identification of switch mode DC–DC power converter based on efficient recursive DCD-IIR adaptive filter, IEEE Transactions on Power Electronics 27 (2012), 4425–4435.

143.

Ahmeid

, Armstrong

, Gadoue

, Al-Greer

and Missailidis

, Real-time parameter estimation of dc–dc converters using a self-tuned Kalman filter, IEEE Transactions on Power Electronics 32 (2016), 5666–5674.

144.

Kardan

M.A.

, Asemani

M.H.

, Khayatian

, Vafamand

, Khooban

M.H.

, Dragičević

, et al., Improved stabilization of nonlinear DC microgrids: Cubature Kalman filter approach, IEEE Transactions on Industry Applications 54 (2018), 5104–5112.

145.

Poon

, Jain

, Spanos

, Panda

S.K.

and Sanders

S.R.

, Fault prognosis for power electronics systems using adaptive parameter identification, IEEE Transactions on Industry Applications 53 (2017), 2862–2870.

146.

and Low

, Low sampling rate online parameters monitoring of DC–DC converters for predictive-maintenance using biogeography-based optimization, IEEE Transactions on Power Electronics 31 (2015), 2870–2879.

147.

Peng

and Wang

, Application of digital twin concept in condition monitoring for dc-dc converter, in 2019 IEEE Energy Conversion Congress and Exposition (ECCE), 2019, pp. 2199–2204.

148.

Milton

, De La O

, Ginn

H.L.

and Benigni

, Controller-embeddable probabilistic real-time digital twins for power electronic converter diagnostics, IEEE Transactions on Power Electronics 35 (2020), 9852–9866.