Hexadrive Engineering: Digital Twin Software for Electrical Machines
by Francesco Toso, Silverio Bolognani, Piergiorgio Alotto, Milo De Soricellis, Riccardo Torchio from Hexadrive Engineering, Spin-off of University of Padua
A digital twin is a real-time replica of a physical system based on a defined mathematical model. It is a soft-sensor tool for monitoring and predicting the behaviour of a single device or multiple interconnected components. Recent advances in computational pipelines, multi-physics solvers, artificial intelligence, big data cybernetics, and data processing and management tools are bringing the promise of digital twins and their impact on society closer to reality . The digital twins’ market growth is driven by Industry 4.0 transformation and by the demand for enhanced embedded software solutions, with estimated global revenues that reach $9.4 Billion by 2025. General Electric and Siemens are the leaders with a revenue growth of 30% in 2019. The most relevant market fields for digital twins are the automotive and manufacturing markets.
Digital twin for the automotive sector
The Electric Vehicle (EV) global market will grow 10 times in the next 10 years and the growth speed will double in the next 20 years. Technology advances are decreasing manufacturing costs, international policies are facilitating and incentivizing EV expansion, hence EV production and vehicle systems are facing scalability challenges. In the automotive industry, a digital twin is now being perceived as one of the biggest disruptors in recent times. The value proposition of the technology proposed by Hexadrive is manifold, and it can be recognized in three different stages: testing, monitoring and data collection. Indeed, a digital twin of the e-traction system will allow complete simulation of the system in all possible scenarios, without spending energy and breaking expensive components. It will allow the execution of failure monitoring of the e-traction system in real-time, detecting errors in production and running diagnostics. It allows for recording event data and preventing dangerous conditions, which ultimately increases safety and reliability at scale. Furthermore, it permits the collection of system data for components’ lifecycle and upkeep, which aids in predictive maintenance and operational planning as well as new feature development.
Digital twin for manufacturing
The industry is driven toward sustainability. Investing in the efficiency and digitalization of production plants is becoming a valuable asset and an effective mean to improve the profitability of the industrial facility. Electrical machines that consume the most power in the industrial field often run at low efficiency during long-term operation. When an electrical motor stops, costs start running, and the cost of repairing the motor is the smallest one. The bigger cost is due to lost production, regulatory fines, and lost reputation. By using digital twins of the most powerful motors present in the machinery of the plant, you can predict failures in advance and reduce downtime, better manage spare part inventories, monitor and manage your fleet, do what-if simulations, and optimize operations. Hexadrive proposes to introduce digital twins of the most powerful motors present in the machinery of the plant. This may lead to the following benefits: 1) monitor in real-time the condition of the machines; 2) run diagnostics and send continuous feedback on its state; 3) record data, adapt and optimize the operating condition; 4) apply predictive maintenance estimating the Remaining Useful Life (RUL); 5) plan optimized cost-efficiency operational of the machinery. Then, all the digital twins are connected to each other exchanging information based on fault identification protocols with different levels. This will help with the correct timing for maintenance and repair actions initiated by the diagnostic system, allowing the motors to last longer at the lowest cost, without suffering quality problems, accidents, or serious failures.
Hexadrive Engineering technology
Our approach consists in using highly accurate numerical and analytical models to create digital twins for model-based condition monitoring. The process of designing our digital twins consists of the following main steps:
- Design of a high-fidelity mathematical model of the physical asset;
- Order reduction of the high-fidelity model ;
- Training of the reduced model by means of simulations and experiments cross-play data analysis;
- Design of prediction models.
Thus, we propose three pillars that the industry should consider in order to exploit the full potential of digital twins:
• Virtual Twin: Creation of a virtual representation of a physical asset;
• Predictive Twin: Physics-based, data-based, or hybrid models to predict the behaviour of the physical asset;
• Improved Twin: Data analysis to improve decision-making and to build new models.
The launching of the Industry 4.0 program contributed to momentum in the adoption of the Digital Twins, being one of the digital disruptive technologies of the upcoming years. This technology has evolved from the simple virtualization into a complex digital copy of real assets, being characterized by the combination of several technologies, such as optimization, real-time data and machine learning.
 Rasheed, Adil, Omer San, and Trond Kvamsdal (Jan. 2020). “Digital Twin: Values, Challenges and Enablers from a Modeling Perspective”. In: IEEE Access PP, pp. 1–1. doi: 10.1109/ACCESS.2020.2970143.
 Hartmann, Dirk, Matthias Herz, and Utz Wever (2018). “Model Order Reduction a Key Technology for Digital Twins”.