Unveiling Innovative Energy Solutions by Digital Twin Technology |
|
|
The MesH Engineering Team is pleased to extend an invitation to you for our Forum at Husum Wind 2023, where we will be showcasing the revolutionary “Renewable Energy communication network (MesH REcon)” that seamlessly integrates energy generation and storage systems such as PV panels, wind turbines, and batteries, while also incorporating dynamic digital twins of these systems, across multiple operational hierarchical levels. This innovation is poised to reshape the energy sector and beyond.
Date: September 15th, 2023, 11:00 – 11:50 am Location: Husum Wind Fair, Forum Hall 2
During the trade fair forum, we will:
We firmly believe that e-TWINS technology has the potential to make a meaningful contribution to advancing the industry by enhancing the efficiency, sustainability, and flexibility of energy generation systems.
We are excited to welcome you to the trade fair forum and explore the innovative potentials of the e-TWINS iniative together. |
Moderation: Dr. Birger Luhmann MesH Engineering GmbH
Speakers: Anton Kaifel Zentrum für Sonnenenergie- und Wasserstoff-Forschung (ZSW)
Jonas Petzschmann Zentrum für Sonnenenergie- und Wasserstoff-Forschung (ZSW)
Abhinav Anand Technical University of Munich
Andre Thommessen University of Applied Sciences Munich
Stefan Hauptmann MesH Engineering GmbH
|
Scenario: Windfarm Control
Goal: Optimizing wind farm operation
We want to operate a cluster of wind turbines as optimal as possible considering the mutual effects of
the turbines (mainly wakes).
In dependency on the influencing factors (e.g. electricity price, inflow conditions, turbine state)
different optimization objectives are formulated including:
- Maximizing the overall power output
- Managing the lifetime considering the structural loading of all wind turbines
- Providing ancillary services for the electricity grid
Problem: Little information on the wind conditions inside the wind farm
Except the turbines itself, measurements of the wind conditions between the turbines are not available.
This leads to uncertainties, e.g. location of the wake and magnitude of the velocity deficit, wind
speed, turbulence.
Reducing the uncertainties is crucial in order to eventually achieve an optimized operational state of
the entire wind farm compared to a modus where the turbines are operated individually.
Our approach: Digital twin technology for enabling wind farm control
With the latest results from research, we have developed a digital twin that calculates the flow
conditions inside a wind farm based on the data from the turbines.
The digital twins are deployed on our e-TWINS-Platform and can be adapted to any wind farm of
choice.
Tested Technology
With our partners from industry and academia we have tested the digital twin for wind farm flows within
the scope of the e-TWINS project.
Scenario: Economic Dispatch
Goal: Economic dispatch and operation of renewable generation/storage plants
To ensure continuous equilibrium between electricity production and consumption at all times, the energy
industry is organised in balancing groups. Each entry and exit of electricity from the grid must be
assigned to a specific group. The so-called balancing group managers coordinates the supply and demand
through aggregation and corresponding market transactions before the point of delivery.
- Higher revenue due to increased market participation
- Guaranteed revenue due to more accurate forecasts
- Reduced penalties/grid service requirements due to feasible power plant generation set-points
Problem: Poor forecasting, reduced utilization of market
When marketing electricity from renewable energy plants, the challenge is that the electricity has to be
marketed in advance, e.g. on the day-ahead market, with forecast uncertainties due to weather-dependent
generation. Some of the forecast deviations that occur can be corrected through transactions on the
short-term markets. However, very short-term changes in the forecast can only be compensated by active
control of the plant capacities.
Our approach: Economic dispatch optimization
In the Economic Dispatch scenario, the optimal marketing strategy is sought for a plant pool consisting
of weather-dependent renewable energy plants and flexible battery storage. The battery storage systems
are used for arbitrage trading as well as for short-term balancing by counteracting forecast deviations.
For the staggered marketing options in the day-ahead and intraday market, the most recent status and
forecast data of the plants are used to determine the optimal power plant deployment. The uncertainty of
the forecasts is incorporated into the optimisation via probabilistic forecasts. At operating time, the
optimal setpoints are first distributed to the plants. The plant operator then ensures that the
setpoints are met based on live data feeds from units and components.
Role of digital twin technology
Digital twin technology serves a vital role in energy systems by enabling continuous communication
between forecasters, balancing group optimizers, and plant control systems (SCADA). It enhances accuracy
through the inclusion of site-specific data, including historical and generation data. Fully modular
digital twins adressing innate hierarchy within the energy system.
We are using digital twins at grid and plant levels to:
- Obtain probabilistic generation forecast
- Stochastic optimization for generating dispatch set-points
- Run dispatch models for flexibilities
- Revision of market bids
- Optimal power plant operation
Scenario: Grid Inertia
Goal: Inertia emulation and grid-forming capability
Goal: Inertia emulation
We want to enable reliable and smart operation of the future power system with high share of renewable
energy systems, such as wind turbine systems. Due to less synchronous rotating masses, the future power
system requires inertia emulation for inverter-based resources.
New inverter control methods with inertia emulation and grid-forming capability support the grid
frequency and voltage. Power curtailment strategies may provide additional power reserves. Monitoring
and forecasting the grid support capability is crucial to maximize the renewable power generation while
ensuring grid stability.
Problem: Decreasing power system inertia and grid-forming capability
More and more inverter-based resources replace directly grid-connected synchronous machines. Latter
provide power system inertia and form the grid voltage in contrast to the standard grid-forming inverter
control. At the same time, the installation of additional transmission capacities, such as HVDC systems,
leads to potentially higher power imbalances during so-called system split scenarios. Both, the
decreasing inertia and the increasing worst-case power imbalance, are major challenges for the future
grid stability.
Our approach: Digital twins for inertia monitoring and forecasting
With the latest results from research, we have developed services based on digital twins that report the
grid support capability of wind and battery units. For wind units, the operating point depends on the
wind speed and the power setpoint. Moreover, the operating constraints, such as power, torque, speed
limits, are taken into account to evaluate wind unit simulations for the worst-case Rate of Change of
Frequency (RoCoF event). The numerical optimization algorithm iterates with different virtual inertia
constant parameter H_v to find its maximum feasible value H_(v,max), reported to higher levels via the
Reporting Module. Thus, based on the inertia monitoring and forecasting of all generation units, the
grid operator can determine the actual virtual inertia demand and set the virtual inertia constant
reference H_(v,ref) for the real Control Module.
The digital twins are deployed on our e-TWINS-Platform and can be adapted to any wind farm of choice.
Tested Technology
With our partners from industry and academia we have tested the grid-forming control methods, the power
curtailment strategies and the digital twins for inertia monitoring and forecasting within the scope of
the e-TWINS project.
Scenario: Virtual Sensor
Goal: Online estimation of loads without measuring it
We introduce a cutting-edge software module that takes condition and structural health monitoring
systems to the next level. This innovation seamlessly becomes part of the digital twin's framework,
enhancing its capabilities by enabling real-time assessments. With precision, it identifies the complex
loads acting on different mechanical components within the drivetrain. This insightful analysis comes
from a smart combination of various measurements, crafting a complete reflection of what's happening
beneath the surface.
Problem: Sensors cannot be feasibly installed in the required positions
The feasibility of installing load sensors and strain gauges in critical positions and hotspots within
the drivetrain can sometimes present challenges. While the idea of having these sensors to gather
precise data is attractive, practical constraints often come into play. These constraints encompass
issues such as limited space for sensor placement, environmental factors affecting sensor performance,
and the financial considerations of sensor integration, making it imperative to strike a balance between
data precision and the practicality of implementation.
Our approach: Data-driven algorithm based on profound knowledge of the underlying physical model
We've developed an innovative data-driven algorithm that combines the insights from physics-based
multibody simulation with the adaptability of data-driven and machine learning techniques. This synergy
leverages the diverse nature of available data to create what we call a "loads virtual sensor." This
sensor harnesses the power of both worlds – the deep understanding rooted in physics and the dynamic
capabilities of data-driven methodologies.
Tested Technology
The technology we've developed has undergone rigorous validation across various operational scenarios,
consistently showcasing remarkable proficiency in accurately estimating both loads and potential damage
incurred by drivetrain components. This validation procedure has been executed in several operational
conditions that the wind turbines suffer from according to widely adopted industrial codes and standards
Renewable Energy communication network
Goal: Developing the Renewable Energy communication network (MesH REcon)
Design, implementation, application and demonstration of a software platform to support the holistic application of digital twins to the future energy system under consideration of scalability, modularity, expandability and transfer to large supra-regional systems.
Background: Digital Twin Systems
A Digital Twin System includes five essential elements:
The Physical Asset (1) represents the real-world counterpart that can range from individual components to entire power plants or grids. It also includes systems like control and SCADA/FPGA/DSP for interfacing with the Digital Twin.
The Virtual Model (2) runs in tandem with the physical asset, serving as a digital replica. It generates virtual measurements that facilitate tasks such as fault detection and optimization.
Data (3) plays a crucial role, encompassing real and virtual measurements, historical data and predictive information. Data management ensures seamless data exchange among different parts of the Digital Twin System.
Services (4) enhance the Digital Twin System by providing a range of functionalities. These services are categorized into two types: intrinsic services, which are essential for the Digital Twin System's operation (such as data coordination), and extrinsic services, which offer additional capabilities (like condition monitoring).
Connections (5)serve as communication channels within the Digital Twin System, enabling the exchange of data and commands among physical assets, data, virtual models, and services.
In summary, Digital Twins merge physical assets with their digital counterparts, utilizing data, services, and connections to optimize performance and support decision-making processes.
Problem: The interconnected nature of complex power systems
The System of Digital Twin Systems is a concept developed to address the intricate and interconnected nature of complex (power-)systems. It incorporates several vital components:
It establishes a clear Hierarchy with four distinct system levels: Component, Unit, Plant, and Grid (System) Level. Each of these levels includes physical assets and their respective Digital Twins. To ensure effective communication and coordination, Interconnections play a crucial role, encompassing both horizontal links for data exchange and automated control actions at the same level and vertical links that capture hierarchical dependencies between different levels.
The fundamental building block of the System of Digital Twin Systems is the Digital Twin System, which consists of a physical asset, a virtual model, data, services, and internal connections. Control and data flow within the System of Digital Twin Systems are facilitated through bidirectional connections, enabling the exchange of commands and measurements between the physical asset and its digital twin.
Vertical Interconnections further enhance coordination and data sharing across various system levels. The cascading structure of the System of Digital Twin Systems reflects the natural hierarchy of complex systems, where higher-level assets, such as wind farms, are composed of multiple lower-level assets like individual wind turbines. Direct communication capabilities allow Digital Twins at the same or different levels to interact efficiently, reducing latencies and supporting rapid information exchange.
Finally, the System of Digital Twin Systems adopts a Modular Approach, making it adaptable and extendable without the need for extensive modifications, as it builds upon individual Digital Twin Systems.
In summary, the System of Digital Twin Systems offers a comprehensive framework for representing and managing complex systems, organizing physical assets and their Digital Twins into hierarchical levels, and enabling efficient interconnections, control, and data exchange.
Our approach: MesH REcon software framework
To implement the System of Digital Twin Systems described earlier, the software framework MesH REcon is proposed. MesH REcon aims to enable seamless data handling among multiple clients, including physical assets, virtual models, services, and local datasets. To meet the requirements of the future power system, MesH REcon should provide functionalities such as data exchange, data persistence, data compatibility, cloud-based access, data authorization, data aggregation and more.
The core idea of MesH REcon is to create a cloud-based platform that serves as a central communication hub and database for DTs and their corresponding physical assets. It stores data in nested key-value structures called Things and allows for communication and data exchange between connected clients.
Additionally, suitable libraries for connecting models through various protocols are available in common programming languages. However, to fully meet the requirements of the MesH REcon, additional features are needed to connect physical assets to MesH REcon. To enable the exchange of virtual dynamic models among different software platforms and co-simulations among stakeholders, an interface between MesH REcon and functional mock-up interface (FMI)-based models must be developed.
MesH REcon´s structure involves connecting a single Digital Twin System as an example, demonstrating interactions, communication and data flow. Intrinsic services, such as the history service for storing and accessing time historic data and the aggregation service for automatic data grouping and calculations, are provided by MesH REcon.
Interactions, data exchange and service configurations occur through the same entry point and APIs for all connected clients. Digital Twin functionalities like Machine Learning can be implemented locally as external services. MesH REcon acts as a communication hub with data integration, while computing tasks are performed locally at each client, optimizing distributed (edge) computing capacity to conserve data transmission bandwidth.
Tested Technology
With our partners from industry and academia we have tested the REcon platform and its digital twin technology in scenarios like wind farm flow, economic dispatch, and System Inertia Monitoring within the scope of the e-TWINS project.
Welcome to the MesH Renewable Energy communication network!
MesH REcon is our premier platform designed to drive digital transformation within the sustainable energy sector. Engineered specifically for the seamless deployment of digital twins, it serves as the nexus where virtual and physical energy elements converge,raising unparalleled connectivity and synergy. This reflects our resolute commitment to pioneering innovation and sustainability in renewable energy.
In plant operation and monitoring, digital twins revolutionize efficiency by accurately representing flow fields within wind parks and predicting local load and stress situations at turbine components.
For Economic Dispatching, we use the power of digital twins to forecast renewable energy output and optimize generation strategies in real-time, ensuring the most economically efficient approach.
As the energy landscape evolves with the rise of inverter-based resources, ensuring grid stability becomes vital. Digital twins play a key role by enabling renewable energy sources to efficiently respond to demands, thereby optimizing their economic performance and boosting grid stability, even within frequency disturbances.
Digital twins empower renewable energy sources to efficiently respond to demands from Redispatch 2.0, optimizing their economic performance.
Through these applications and more, the MesH Renewable Energy communication network and digital twins pave the way for a future where sustainability and technological advancement harmoniously coexist, driving positive change in the renewable energy landscape.
For more information, please use our interactive chart.
