DIGITAL TWINS
Digital Twins for Energy Production
Digital Twin of the Microgrid with Forecast-Driven Control
Digital Twin of the Production Process with Forecast-Driven Control
Digital Twin for Top-Level Orchestration
Virtual lab – Test Your EMS with a 10 MW Microgrid Directly on Your Desk
Detailed Microgrid Modelling and CAPEX Optimization
Digital Twin for Predictive Control of Electric Vehicle Drivetrain Inverter
Digital Twin for Predictive Trajectory Estimation for Ecological Route Planning
Multilevel Digital Twin for State-of-Health (SoH) Estimation of Drivetrain Inverter
Digital Twin for Real-Time Emulation of Electric Vehicle Powertrain in a Power-HIL Environment
Real-time Digital Twin for Hardware-in-the-Loop (HIL) or Power-HIL testing
Digital Twins for Power Electronics
Advanced Wide-Bandgap (SiC & GaN) Topology Modelling
High-Power Thyristor & IGBT Converter Simulations
Deep Electro-Thermal Domain Integration
High-Fidelity Transient & Switching Analysis
Reliability & Degradation Modelling
Real-time Digital Twin for Hardware-in-the-Loop (HIL) or Power-HIL testing
Digital Twins for Production Process
Digital Twins for Predictive Control
Synthesized Data for Machine Learning
Digital Twins for Predictive Maintenance
Digital Twins for Hardware-in-the-Loop (HIL) testing
DIGITAL TWINS FOR ENERGY PRODUCTION
MW-Engineering is a member of the OpenEMS Association e.V., specializing in the development of advanced Digital Twin systems for energy and industrial applications. Our solutions are designed to provide measurable improvements in performance, reliability, and economic efficiency. Below is an overview of several Digital Twin–based capabilities that can significantly strengthen the value, functionality, and competitiveness of your projects.
1. Digital Twin of the Microgrid with Forecast-Driven Control
A predictive digital twin incorporating solar and electricity price forecasts enables dynamic BESS dispatch strategies. This maximizes commercial benefit, and shortens the payback period of the system.
2. Digital Twin of the Production Process with Forecast-Driven Control
A high-fidelity Digital Twin of the production process, combined with solar generation and electricity price forecasts, enables dynamic optimization of manufacturing operations. The system identifies the most cost-efficient production schedule, reduces grid electricity consumption, and increases the utilization of on-site solar energy. When integrated with a BESS, the Digital Twin also generates an optimal BESS dispatch strategy to further decrease production costs and improve overall energy efficiency.
3. Digital Twin for Top-Level Orchestration
Having distributed network of clients, featuring various capacity combinations and microgrid architectures, the digital twin is an excellent fit for use within the Orchestrator - top-level management system. The digital twin allows for forecasting the behavior of system participants and establishing preemptive control strategies without waiting for critical events.
4. Virtual lab – Test Your EMS with a 10 MW Microgrid Directly on Your Desk
It is possible to test an EMS without any physical microgrid hardware. The EMS is tested using a Hardware-in-the-Loop approach, where the EMS is connected through standard interfaces exactly as in a real microgrid, while a full real-time simulation of the microgrid operates in place of the physical system.The setup is fully scalable and allows EMS validation under different microgrid configurations and power levels, across a wide range of operating modes, including fault and emergency scenarios that cannot be tested safely in practice.This virtual lab can also be used for testing load balancers for EV charging or other control systems requiring realistic real-time behavior.
5. Detailed Microgrid Modelling and CAPEX Optimization
High-fidelity modelling of the microgrid enables accurate sizing of all components (BESS capacity, inverter power, PV, gensets, chargers, etc.) and precise definition of EMS operating modes. This approach allows CAPEX optimization of up to 20%. Additionally, it makes it possible to evaluate system behavior in the exact use-case, assess cycle count, and analyze operational constraints such as minimum/maximum SoC and stop/start conditions.
DIGITAL TWINS FOR AUTOMOTIVE
1. Digital Twin for Predictive Control of Electric Vehicle Drivetrain Inverter
A predictive digital twin of the drivetrain inverter enables advanced Predictive Control strategies. By forecasting load demands and thermal states, the system optimizes inverter states in real-time. This approach minimizes power losses and reduces thermal stress, thereby enhancing reliability and extending service life. Ultimately, this increases overall system efficiency and sustainability.
2. Digital Twin for Predictive Trajectory Estimation for Ecological Route Planning
Integrating vehicle dynamics with route topography and traffic data, this digital twin accurately predicts energy consumption for various route options. It enables "Eco-routing" and optimal velocity planning to minimize battery usage. Similar to production optimization in energy systems, this maximizes the utilization of available on-board energy and extends vehicle autonomy.
3. Multilevel Digital Twin for State-of-Health (SoH) Estimation of Drivetrain Inverter
A high-fidelity, multilevel digital twin—ranging from chip thermal models to system-level behavior—provides precise monitoring of the inverter's State-of-Health (SoH). This system enables accurate estimation of Remaining Useful Life (RUL) and facilitates predictive maintenance strategies, preventing critical component failures and optimizing the lifecycle cost of the drivetrain.
4. Digital Twin for Real-Time Emulation of Electric Vehicle Powertrain in a Power-HIL Environment
This solution provides a real-time, high-fidelity emulation of the entire EV powertrain for use in Power-Hardware-in-the-Loop (P-HIL) setups. It allows for the testing of physical hardware components (e.g., drivetrain inverter) under realistic electrical and mechanical loads in a laboratory setting. The system supports validation against standardized drive cycles (e.g., WLTP, HWFET, US06) and extends capabilities up to Driver-in-the-Loop implementations, allowing real-time human interaction with the simulated vehicle. This eliminates the need for full vehicle prototypes during early development stages.
5. Real-time Digital Twin for Hardware-in-the-Loop (HIL) or Power-HIL testing
Development of high-fidelity Digital Twins optimized for real-time execution, specifically tailored for integration into Hardware-in-the-Loop (HIL) and Power-HIL environments. This approach unlocks the full value of HIL testing by enabling validation in a safe, repeatable, and controlled virtual setting. It allows for the rigorous assessment of system behavior under extreme fault conditions and emergency scenarios that are too dangerous or costly to perform on the road, significantly accelerating validation cycles and ensuring control system robustness.
DIGITAL TWINS FOR POWER ELECTRONICS
1. Advanced Wide-Bandgap (SiC & GaN) Topology Modelling
We provide high-fidelity models for modern converter topologies utilizing Silicon Carbide (SiC) and Gallium Nitride (GaN) technologies. This includes B6 SiC MOSFET bridges for EV drivetrain inverters and Active Front Ends (AFE), as well as Interleaved Single-Phase PFCs based on GaN. The digital twins accurately reflect the behavior of these fast-switching devices, enabling precise design of high-efficiency power stages.
2. High-Power Thyristor & IGBT Converter Simulations
Drawing on extensive industrial experience, we offer detailed digital twins for heavy-duty power conversion. This covers classic 6-pulse and 12-pulse thyristor rectifiers and high-power IGBT inverters. These models are essential for analyzing grid interactions, harmonic distortion, and control stability in large-scale industrial drive systems and grid-tie applications.
3. Deep Electro-Thermal Domain Integration
Our digital twins go beyond simple electrical simulation by incorporating a coupled electro-thermal domain. We model both static (conduction) and dynamic (switching) losses to calculate the Junction Temperature (Tj) in real-time. This allows for the precise dimensioning of cooling systems and the analysis of thermal stress under varying load profiles.
4. High-Fidelity Transient & Switching Analysis
For critical component validation, the digital twin captures fast transient phenomena at the physical level. This includes the modeling of reverse recovery currents, parasitic inductances, and voltage overshoots during switching events. This level of detail is crucial for optimizing gate drivers, designing snubber circuits, and ensuring electromagnetic compatibility (EMC).
5. Reliability & Degradation Modelling
To support predictive maintenance, our models track Thermperature-Sensitive Electrical Parameters (TSEPs) and degradation-sensitive parameters (DSEPs). By monitoring shifts in TSEPs, the digital twin can estimate the aging of semiconductor modules and predict end-of-life scenarios, enabling a transition from scheduled to condition-based maintenance.
6. Real-time Digital Twin for Hardware-in-the-Loop (HIL) or Power-HIL testing
Development of high-fidelity Digital Twins optimized for real-time execution, specifically tailored for integration into Hardware-in-the-Loop (HIL) and Power-HIL environments. This approach unlocks the full value of HIL testing by enabling validation in a safe, repeatable, and controlled virtual setting. It allows for the rigorous assessment of system behavior under extreme fault conditions and emergency scenarios that are too dangerous or costly to perform on the road, significantly accelerating validation cycles and ensuring control system robustness.
DIGITAL TWINS FOR INDUSTRY
1. Digital Twins for Production Process
Digital Twin serves as the core of a predictive control system, enabling the simulation of thousands of future operational scenarios to identify the most profitable strategy. By integrating real-time forecasts (e.g., energy prices, raw material variability) with strict process constraints, the system automatically generates an optimal production schedule that minimizes unit costs. This proactive approach ensures maximum efficiency and profit margins by adapting to changing external conditions before they impact the process.
2. Digital Twins for Predictive Control
Predicting the behavior of complex dynamic systems characterized by multiple interlocking feedback loops is a major challenge for traditional methods. The Digital Twin effectively solves this by providing a high-fidelity simulation of the physical process. It allows the system to calculate future behavior under various conditions and automatically construct an optimal control strategy based on your specific criteria—whether it is maximizing throughput, minimizing energy costs, or ensuring stability.
3. Synthesized Data for Machine Learning
Real-world data is often insufficient or lacks the critical failure examples required for training robust AI models. We leverage high-fidelity Digital Twins to generate synthesized datasets, effectively solving the "cold start" problem and bridging data gaps. This approach allows you to train ML algorithms on dangerous edge cases and emergency scenarios that are impossible to reproduce physically. Furthermore, it ensures your models are validated across a wide variety of operating conditions and environmental factors.
4. Digital Twins for Predictive Maintenance
Acting as a real-time reference model, the Digital Twin runs in parallel with your physical asset to continuously compare actual telemetry against simulated ideal behavior. This enables the detection of subtle anomalies and early-stage degradation long before they trigger standard alarm thresholds. By integrating physics-based wear models, the system accurately estimates Remaining Useful Life (RUL), allowing you to transition from costly fixed schedules to optimized condition-based maintenance.
6. Real-time Digital Twin for Hardware-in-the-Loop (HIL) or Power-HIL testing
Development of high-fidelity Digital Twins optimized for real-time execution, specifically tailored for integration into Hardware-in-the-Loop (HIL) and Power-HIL environments. This approach unlocks the full value of HIL testing by enabling validation in a safe, repeatable, and controlled virtual setting. It allows for the rigorous assessment of system behavior under extreme fault conditions and emergency scenarios that are too dangerous or costly to perform on the road, significantly accelerating validation cycles and ensuring control system robustness.
PROJECTS
Digital Twin of the titanium sponge full production process
Project completed: 2021.
Results of the Digital Twin implementation:
-
Identified bottlenecks in the technological process and implemented corrective actions in production.
-
Determined parameters for new process equipment, enabling an increase in production line productivity.
Technical challenge:
The production process contains multiple technological feedback loops with variable coefficients, requiring an iterative modeling approach and making it impossible to represent the process as a system of equations.
Therefore, the application of a Digital Twin was essential to accurately reproduce process dynamics, identify dependencies, and support optimization of technological parameters.
Intelligent Diagnostic System for Metallurgy Stand Mill 1700 Based on the Digital Twin Concept
Project completed: 2021.
A real-time State-of-Health monitoring system for the mechanical, electrical, and thermal subsystems of the Metallurgy Stand Mill 1700, combined with predictive maintenance capabilities.
The project included the development of digital twins for 12-pulse thyristor converters (2 × 6300 A, 1050 V) and a 7.1 MW drive motor. Real-time parameter estimation error did not exceed 5%, and the system also assessed motor insulation degradation.
Digital Twin for Predictive State Trajectory Estimation of an Electric Vehicle Drivetrain Inverter
Project completed: 2025.
Digital Twin was developed for predictive state trajectory estimation of an EV drivetrain inverter. The predicted state trajectory was used to compute the optimal reference trajectory for Predictive Control System of the drivetrain system. An article was published based on the results of this work "Ecological Route Planning and Execution with Remaining Useful Lifetime Estimation and Predictive Control of Power Electronics".
Digital Twin for Real-Time Emulation of Electric Vehicle Powertrain in a Power-HIL Environment
Project completed: 2025.
Digital Twin was developed for real-time emulation of an electric vehicle (EV) drivetrain in a Power-Hardware-in-the-Loop (Power-HIL) setup. The system was applied for testing innovative inverter designs and validating control strategies under realistic operating conditions.
The contribution was made within the framework of the AUTOtech.agil project.
Multilevel Digital Twin for State-of-Health (SoH) Estimation of Drivetrain Inverter
Project completed: 2025.
Multilevel Digital Twin was designed to assess the State-of-Health (SoH) of drivetrain inverters, combining physical modeling, data-driven estimation methods, and AI-based analytical tools.
This hybrid approach enabled predictive evaluation of inverter degradation and improved accuracy of remaining useful lifetime (RUL) estimation.
The contribution was made within the framework of the AUTOtech.agil project.
Digital Twin of 7.1 MW DC motor for State-of-Health monitoring
Project completed: 2021.
A high-fidelity real-time Digital Twin for a 7.1 MW DC motor, computing 145 internal temperature parameters with <5% estimation error. The system enabled early detection of insulation degradation, providing high diagnostic value and predictive maintenance capability.
Digital twin of mine hoist system
Project completed: 2020.
The Digital Twin was developed for HIL testing of mine hoist equipment responsible for safety-critical motion control. It enabled the verification of operating modes that are too dangerous to test on real machinery, providing a safe environment for validating and improving control algorithms.
Real-time Digital Twin of a B6 thyristor rectifier designed for early fault detection
Project completed: 2020.
A real-time Digital Twin of a B6 thyristor rectifier that enabled early detection of internal faults which could otherwise lead to converter failure. This approach significantly improved system reliability and reduced the risk of unplanned downtime. This approach was successfully validated in laboratory conditions on a 22 kW thyristor converter.
Real-time Digital Twin for HIL testing with the control board
Project completed: 2018.
The Digital Twin consists of detailed models of the thyristor converter, the DC motor, and the mechanical load. The HIL setup was used to refine and validate control algorithms, enabling safe testing of operating modes that would be dangerous to reproduce on real equipment.
...and various multiphysical simulations, backed by solid engineering experience since 2004.
CONTRIBUTION TO SCIENCE
1. O. Solomakha, V. Afanasenko and I. Kallfass, "Multilevel Digital Twin of Power Electronics based on Degradable Wide-Bandgap Semiconductors for State-of-Health Estimation," 2024 IEEE 9th Southern Power Electronics Conference (SPEC), Brisbane, Australia, 2024, pp. 1-5, doi: 10.1109/SPEC62217.2024.10893280.
https://ieeexplore.ieee.org/document/10893280
2. O. Solomakha, K. Munoz Baron and I. Kallfass, "Digital Twin Approach for Accurate System-Level Simulation of Wide-Bandgap Power-Semiconductors using Temperature Dependent Parameters," PCIM Europe 2023; International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management, Nuremberg, Germany, 2023, pp. 1-6, doi: 10.30420/566091279.
https://ieeexplore.ieee.org/document/10173376
3. O. Solomakha, V. Afanasenko and I. Kallfass, "Online Junction Temperature Estimation of Power Semiconductor Devices using Neural Network and Model-Based Design," IEEE EUROCON 2023 - 20th International Conference on Smart Technologies, Torino, Italy, 2023, pp. 187-192, doi: 10.1109/EUROCON56442.2023.10198878.
https://ieeexplore.ieee.org/document/10198878
4. V. Afanasenko, M. Arzner, C. Hermann, I. Kallfass, S. S. Roge, O. Solomakha, "Ecological Route Planning and Execution with Remaining Useful Lifetime Estimation and Predictive Control of Power Electronics," Architectures and tools for future mobility: our intelligent software revolutionizes daily traffic, Issue 1, 2024, pp. 66-69, doi: 10.18154/RWTH-2025-00773.