Application of Digital Twin to Improve Renewable Energy Source Efficiency
Introduction
Integration of renewable energy sources (RESs) into power systems (PSs) is a significant development affecting numerous issues in specific nations. The issue of reducing CO2 emissions is a global one. Renewable energy source (RES)-based electrification is a critical step in improving the world's environmental predicament. We can identify four major renewable sources using detailed art analyses: wind renewable energy, solar renewable energy, hydro renewable energy, and biofuel (see Figure 1). The substitution of renewable energy sources for traditional electric power plants that use fossil fuels can help address this dilemma.
2. RESs' Digital Twin
Digital transformation aids in decarbonizing the energy supply, reducing reliance on fossil fuels, and integrating renewables into PSs while enhancing resilience. RESs can potentially be deployed to provide electricity and heat to shelter towns. A shelter city is a group of residents (refugees) who have lost their homes due to conflict. The construction of shelter cities causes issues in existing PSs and has an influence on power quality in the impacted power grid. The advancement of digital technologies and digital twins (DT) makes this application more dependable and cost-effective. In 2026, the worldwide DT market is anticipated to be worth $41.77 billion.
Shelter city prosumers can be made via DT implementation designed and adapted for war or post-war PS situations, providing additional advantage to the PS (Figure 5). RES DTs manage their day-to-day operations and optimize performance to boost efficiency and accelerate the EU aim of attaining 50% renewable power use by 2030.
Figure 5: Flowchart for DT implementation in DRES operation.
Several research articles focused on practical concerns and gave a case study: an assessment of a PV inverter integration to a DRES-penetrated microgrid and discussed DT of RES equipment. The flowchart in Figure 5 depicts the process of applying DT in RESs.
The CORDIS Results Pack report highlighted EU-funded initiatives that are creating digital solutions (including DT) to provide a secure and diverse energy supply, increase RES efficiency and resilience, decrease emissions, and give citizens with novel energy services. Vision 2030 on Market Design and System Operation offers ENTSO-E's main drivers for power systems over the next decade: the expansion of renewables (green deal), the push for electrification, the increase in decentralized resources, and digitalization.
The INTERFACE project creates a bridge between transmission and distribution system operators and their consumers, allowing for seamless integration and effective use of renewable energy sources (RESs) in the grid. The FLEXIGRID project seeks ways to ensure the security and dependability of the power grid as it includes increasing volumes of renewable energy sources.
The TwinERGY project offers a first-of-its-kind DT platform that will include the intelligence essential for optimizing energy demand and appropriate RES utilization. TwinERGY strengthens and catalyzes collaborative progress in research, innovation, policy, and market concerns related to demand response, RES integration, and consumer engagement.
Furthermore, these DTs will serve as the foundation for new market structures, allowing for greater utilization of distributed RESs. The act of assigning real, physical embodiments and digital identities, also known as digital twins, is at the heart of digital transformation. Then, DT is used to assist activities such as asset design, development, monitoring, and targeted interoperability between reality and digital representations.
The amount of correlation between the physical item and its DT is crucial because it affects the correctness of the information provided through the DT as well as the dependability of any choice made based on it. The PV sector is rapidly advancing down the road of digital transformation. Basic solitary DTs are already in use across the PV asset lifespan, from design to monitoring to decommissioning and disassembly. The present state of study demonstrates the importance of the suggested issue as well as the necessity to create techniques and suggestions for reducing RESs with DT use. These DTs can also help optimize the shelter city's power supply (Figure 6).
Figure 6 shows the DTs for shelter cities and RESs.
3. TSO and DSO integrated DTs for RESs
Table 1 exhibits the primary DT technologies for integrated energy systems.
Key technologies for the integrated hybrid energy system are listed in Table 1.
Due to the shortcomings of integrated DTs in power systems, efforts are being made to develop new strategies for the actual physical application of DTs in TSOs for RESs.
hybrid systems with multiple physics and information, or the DT of any connected energy system, through integrated modeling. Associative modeling of the doublet information-physical object or coupling modeling of the considered complicated energy lines are examples of this. The hybrid model is obtained by shrinking and streamlining the previously stated physics-information twin;
A thorough knowledge of intricate connecting aspects, which is challenging to directly examine, is necessary for an all-encompassing energy system. This is based on the study and computation of mechanisms and makes use of dynamic model optimization for the unsurveyable object. It further considers actual measurable information;
To get a stable solution for the multivariate heterogeneous model, one must concentrate on the solution complexity as determined by specific DT hybrid model components. The fundamental method involves developing a coherent framework for the solution that completely complies with all other model elements, such as the communication interfaces and actual interactions of the suggested hybrid model. The multi-scale collaborative solution is the next stage. The link between the hybrid and heterogeneous models and the parallel stability solution of twin real-digital objects is also necessary.
Additionally, true holographic mirroring capacity is represented by feedback incorporating actual measurements and multi-scene applications. Both items are capable of interacting with digital reality.
4. Characteristics of DT Application for Best Management of Dispersed Power Systems with Deeply Penetrating RESs
In order to implement DT RESs, certain job types must be chosen, such as dynamic systems with ideal RES operation management circumstances. Long-term-short-term forecast, planning, maintenance, and operational control in real-time mode characterize DT RESs. Comparing various control methods is the primary responsibility of DT RESs (operational and automatic). The quality of the DT RES job definition has a significant impact on how economically efficient the DT solution is implemented. The identification of trustworthy virtual models, or mathematical models, that take into account the dynamics of RES generation and load power grid, is a major challenge during the deployment of DT for RESs.
The mathematical formalization of the processes that are optimized for presentation through the use of pricey software is a feature of current optimization techniques. Nonetheless, there are substantial challenges in the mathematical modeling of systems with intricate temporal and spatial hierarchical structures. The primary ones are the numerous control criteria, the wide dispersion, and the requirement to integrate the tasks of operational (dispatching), automated control, and short-term planning in a timely manner.
The following is a generic representation of the relationships that now exist between the parameters of the control process and the parameters of the system elements in which this process occurs:
𝑦(𝑢)=∑𝑖=1𝑚 1𝑎𝑖∏𝑗=1𝑛𝑢𝛼𝑗𝑖𝑗
where uj is the variable system parameters; m1 is the number of function members; n is the number of variables; y(u) is a generalized technical and economic indicator; and ai, αji are constant coefficients dictated by the attributes of the system.
Expression (1) is the objective function and the exponent y is the optimality criteria in the optimization issues under discussion. It is vital to weigh possibilities and select the optimum solution based on a certain criterion while tackling optimal control challenges. The fundamental variables yb and ub should be used to compare the available possibilities. These values allow for the expression of any variation in the system's state. This may be carried out in this manner.
Now let's mark
𝑦=𝑦∗·𝑦𝑏, 𝑢𝑗=𝑢𝑗*·𝑢𝑗𝑏
where the parameters' relative values are 𝑦∗=𝑦/𝑦𝑏 and 𝑢𝑗*=𝑢𝑗/𝑢𝑗𝑏.
when we change (2) into (1), we get
𝑦∗·𝑦𝑏=∑𝑖=1𝑚 1𝑎𝑖∏𝑗=1𝑛𝑢𝛼𝑗𝑖𝑗*𝑢𝛼𝑗𝑖𝑗𝑏
Applying the same modifications and adding a substitute,
We get the problem record's criteria form:
𝑦∗=∑𝑖=1𝑚 1𝜋𝑖𝑏 ∏𝑗=1𝑛𝑢𝛼𝑗𝑖𝑗*
Keep in mind that the criteria equation (5) for the basic variant will look like this when y = yb.
The similarity criteria in the final equation are normalized to 1. They show how each component or member of the function compares to one another in terms of the optimality criterion.
It is clear from looking at criteria Equation (5) that this allows us to examine the ideal solution on the sensitivity as well as the impact of any variable uj deviating from its optimal value on the value of the optimality criterion. It makes the most sense to do this in relative terms, of course. Such a possibility exists if the best solution's sensitivity is to be achieved in specified units by condition. To recalculate deviations or variations, use expression (2).
Naturally, figuring out the ideal (baseline) values for the system's parameters must come first. You can use any known optimization technique to do this. But since criterion programming optimizes variables based on similarity criteria, it is the best option in this scenario. A method was created that allows variables of a direct issue to be transferred from double variables of criterion programming (similarity criteria).
The experience of applying optimization algorithms in operational control practice demonstrates that it is required to continuously modify the RES's settings in order to attain appreciable efficiency.
The primary component of the DT, the optimality criteria, is compared between the present and optimal values to determine the control and correction of states.
Δ𝐹=𝐹𝑐𝑢𝑟 − 𝐹𝑜
where Δ𝐹 is the value that represents the optimality criterion's difference between the system's current value, Fcur, and its optimal Fo over a certain amount of time in order to govern its circumstances.
It is evident that the full coincidence of Fcur and Fo in genuine technological systems is not feasible for a variety of reasons, including the discreteness of the regulatory parameter change, which makes it occasionally impossible. In dynamic systems like power grids, achieving equality Fcur = Fo necessitates a high intensity of DT, which means that their technical resources are used quickly, their reliability is decreased, and as a result, failures and losses occur—sometimes to the same or even greater extent than the technical-economic effects that are attained as a result of optimization.
The investigation of the optimality criterion's sensitivity and the creation of a suitable zone of insensitivity, within which all system versions are economically equivalent, constitute the overall strategy for tackling this challenge.
Establishing a suitable mathematical model is a crucial first step in the investigation and assessment of the sensitivity of optimal solutions, which primarily dictate the efficiency of the system overall. In order to address the issues of sensitivity analysis of optimal solutions for regulating the circumstances of dynamic systems, it is imperative that we create mathematical models. They need to permit examination and, based on the analysis's findings, interpretation of the model's obtained optimum in relation to an optimum of the actual scheme under investigation.
The correctness of the starting data, the practical application of ideal solutions, and the specification of similarity criteria must all be matched in the criterion models used for this purpose.
The systems under consideration are composed of interrelated individual elements, whose aggregate attributes can be represented by graphs or matrices. In the mathematical models employed in optimum control, the connections between the parameters of the graph's edges and vertices are simplified to a quadratic form. Every element in this form is the product of two distinct variables or the square of one of the variables. It is expressed in the form in matrix form.
𝑓(𝑥)=𝐱𝑡𝐐𝐱
where the rank of the symmetric matrix Q is equal to the rank of the form f(x).
When forming the objective function as a function of the graph's vertices' parameters in mathematical models of optimization problems, the quadratic form must be represented in the canonical form [19], which is when it only consists of the sum of terms with the squares of the variables. As a result, the matrix Q must be transformed into a diagonal, which may always be done by orthogonal transformation. Here, the similarity transformation matrix S is applied to the vector x, transforming it in a way that
𝐐̃=𝐒𝑡𝐐𝐒
where S is the matrix sensitivity and 𝐐̃ is the diagonal matrix with the eigenvalues of the matrix Q on its major diagonal.
Therefore, a technique for evaluating the sensitivity of the eigenvalue-based mathematical model of optimum control must be developed.
The determination of the respective insensitivity zones of the regulated parameters and the validity of the optimality criterion's chosen insensitivity zone both have a substantial impact on the quality of optimal control. The SAC needs to be adjusted to the ever-changing operational circumstances of the systems under consideration.
In order to do this, a program and method for identifying the limits of the area of insensitivity (optimality) of the state control vector components of systems whose mismatch between the optimal and present states is typified by significant losses must be developed.
Nevertheless, they fail to consider the impact of parameter determination accuracy on the similarity criteria calculation accuracy as feedback coefficients of the species control laws. Errors in establishing the similarity criteria for control result in an imperfect match between the ideal circumstances of the system's mathematical model and the actual system. In order to form and practically apply the laws of optimal control for the conditions of a dynamic system, it is necessary to develop methods, algorithms, and programs for DT RESs that take sensitivity into account and stray from computational errors and errors in the initial information. DT can assist in including high-penetration RESs in unbalanced power systems through established techniques, algorithms, and programs.
5. The Unbalance Electric Power Systems with RESs
Electric power networks are found to be unbalanced, and not just in Ukraine. This phenomena can be attributed to the exponential growth in PPP and WPP generating. Inadequate ability to maneuver and exert enough power to maintain equilibrium might also result in an imbalance across the system.
In light of the yearly trends of increasing installed capacity and power generation from renewable energy sources (RESs), the current state of the electric networks presents new problems and difficulties.
Unbalanced electrical networks are currently a very serious problem in Ukraine. The established norm for financial accountability for power imbalances in the electrical system has been in place since January 1, 2021.
Companies that provide energy using renewable energy sources (RESs) will be required to pay financial levies for power imbalances. Under these circumstances, the issue of RES instability analysis in electricity generation under the management of electric network mode parameters becomes real [24]. Owners of PPS, WPS, and mini-HPS systems must estimate power generation with the least amount of inaccuracy feasible in order to reduce the possibility of having to make up for wrong forecast data. For the cost-effective integration of renewable energy sources (RES) including wind, solar, and hydro into microgrids, local and regional distribution grids, and national transmission networks, forecasting and the use of DTs are becoming important techniques.
It should be noted that the issue of reliable and accurate forecasting still requires new in-depth studies and research projects, despite the abundance of software packages and algorithms that make forecast data formation possible. This is because the forecasting process is made much more difficult by the ongoing changes in the local weather and climate. This article is devoted to the study of the instability of RES generation in the regulation of electrical networks. The forecasting process comes before the analysis of the instability of RES generation, namely the identification of the most significant meteorological elements.
It should be highlighted that despite the abundance of software and algorithms that enable the creation of forecast data, the problem of trustworthy and accurate forecasting still necessitates careful investigation and study because weather patterns are constantly changing, greatly complicating the forecasting process.
This article is devoted to the study of the instability of RES generation in the regulation of electrical networks. The forecasting process comes before the analysis of the instability of RES generation, namely the identification of the most significant meteorological elements. Thus, Figure 7 depicts characteristics of operational electricity systems in contemporary times.
Figure 7. Characteristics of operational electricity systems in contemporary times
Under RES integration, it is feasible to ensure balance and, consequently, improved dependability in the electricity system by using the methods and strategies shown in Figure 8.
Figure 8: Strategies for maintaining power system balance reliability when high-grade RES integration is present.
0 Comments