Doctoral Researcher / Master's Thesis Worker - VIcTOR - Virtual Torque Sensing for Maritime Powertrains

Application closes on

Unit

Job category

The School of Engineering drives science and innovation in industrial and built environment technologies. We are committed to educating a new generation of experts who combine technical excellence with a deep understanding of sustainable development in shaping societies. Our research focuses on sustainable built environment, mechanics and materials, multidisciplinary energy technology, and design and implementation of technical systems. The strength of our school lies in close collaboration with stakeholders across research and education. About 45 doctoral candidates and 350 master’s students graduate from the school every year. The school is home to 700 staff members, including 70 professors. To learn more, please visit eng.aalto.fi.

School of Engineering is an international unit with nearly 70 talented professors and 3400 full-time students. Currently, the Department of Mechanical Engineering has 30 professors together with lecturers, post-doctoral researchers, doctoral students, and technical staff. In total, the number of personnel is about 300.

Aalto University Rotor Lab at the Department of Mechanical Engineering, in collaboration with our industrial partner Wärtsilä, is looking for a

VIcTOR - Virtual Torque Sensing for Maritime Powertrains

Description of the project

Knowledge of the torque and torsional response of a maritime powertrain is important for emission-compliant control. Torque and torsional response are normally measured using strain gauges, dedicated torque sensors, or other instrumentation installed on the drivetrain. Torsional response can also be calculated if the external torsional forces acting on the powertrain are known, but measurements of this external forcing are rare in maritime powertrains. Instrumenting large maritime powertrains is also costly, and there is often not enough measured information available to fully determine the torsional response using conventional methods alone.

Unknown external forces and torsional response can instead be estimated using methods such as convex optimization or Kalman filtering. These torque estimation methods allow much more freedom in sensor placement and reduce the number of sensors needed to characterize the full torsional response of the powertrain.

This project aims to develop a torque estimation procedure that determines the unknown input torque and full torsional response using only encoder (speed, position) measurements together with selected available measurements from the driving internal combustion engine (e.g. cylinder pressure) or the driven electric generator (e.g. electromagnetic torque) — removing the need for an expensive, dedicated torque transducer.

Two complementary approaches will be explored. The first is convex optimization-based torque estimation, which allows constraints to be incorporated flexibly, both to keep estimated quantities physically consistent and to enforce desired structure in the results. The second is physics-informed learning, which combines available historical data with known system dynamics as a regulariser for data-driven learning — enabling nonlinear system dynamics to be learned indirectly with less training data than traditional data-driven approaches, for example to estimate the external torsional excitation from the propeller.

To ultimately remove the need for a torque flange, the estimation method must be integrated into the embedded engine control system. The work will make use of industry-standard tools such as Matlab Simulink to connect with existing control systems and support SIL/MIL simulation, and the resulting algorithm must run efficiently within the CPU constraints of an embedded system.

The project is carried out in collaboration with Wärtsilä, which provides powertrain data, operational torsional-dynamics measurements for model verification, and access to a physical maritime powertrain for validating the developed torque estimation procedure. You will be employed by the University, but your work will be conducted physically both at Aalto University campus and Wärtsilä Vaasa STH laboratory, to provide best-possible methodological support for your doctoral dissertation project.

Research focus

• Defining the accuracy and operational requirements for the torque estimation algorithm
• Building and verifying a torsional vibration model of the powertrain, based on data provided by Wärtsilä, with a final test environment in Simulink
• Developing convex optimization-based and/or physics-informed deep learning approaches to torque estimation
• Studying the torque estimation framework first with simulated measurements and scenarios, then verifying it against measurements from Wärtsilä's physical powertrain
• Testing and refining the torque estimation model for implementation in an embedded environment with limited computing capacity

Skills we are looking for

Ready to apply?

If you want to join our community, please submit your application through our recruitment system by November 11th 2026 at 23:59Finnish time (UTC +2).

Please note: Aalto University's employees and visitors should apply for the position via our internal HR system Workday (Internal Jobs) using their existing Workday user account (not via the external webpage for open positions). Aalto University's students and visitors should apply as external candidates with a personal (not aalto.fi) email.

The position will be filled as soon as a suitable candidate is identified.

Application materials (all in English, compiled into one PDF)

• Motivation letter with contact information and preferred starting date (max. 1 page)
• CV including academic merits (max. 3 pages) and contact details of two referees
• Copy of most recent thesis (Bachelor's or Master's)
• Degree certificate(s) and transcript of records with explanation of grading scale

Please note that these are considered assets, not strict requirements:

• Background in mechanical engineering, control engineering, applied mathematics, electrical engineering, or a related field
• Knowledge of torsional vibration, rotordynamics, or powertrain/drivetrain dynamics
• Experience with state estimation or optimization methods (e.g. Kalman filtering, convex optimization)
• Interest or experience in machine learning, particularly physics-informed or physics-constrained learning
• Programming and modelling skills (Matlab/Simulink, Python, C or other embedded-systems languages)
• Interest in embedded systems and control engineering, and in working with industrial measurement data

General eligibility criteria

• Master's or Bachelor's degree in mechanical engineering, or other relevant field
• Excellent written and oral skills in English
• Finnish language skills are considered a benefit
• Applicants must be able to work full-time and, for doctoral studies, meet the official eligibility requirements of the Aalto Doctoral Programme in Engineering during a trial period of 6 months: aalto.fi/en/study-options/aalto-doctoral-programme-in-engineering

What we offer

• Salary during Master's thesis phase: 2622 €/month (gross), full-time, max. 6 months, employment at Wärtsilä
• After graduation, doctoral researcher starting salary: 3143 €/month (gross), with possible increases based on achievements
• Annual workload: 1612 hours
• Employment contract includes occupational health care; Finland has a comprehensive social security system
• First doctoral contract: 2 years, followed by a midterm evaluation and extension for up to 4 years in total
• Strong collaboration with Wärtsilä (access to engines, test rigs, operational data, and industry experts)
• International and multidisciplinary research environment with excellent facilities

Further information

For additional information, please contact:Assistant Professor Raine Viitala (raine.viitala@aalto.fi)

For recruitment system questions, please contact: hr-eng@aalto.fi

More about working at Aalto: aalto.fi/en/careers-at-aalto

Virtual campus tour: virtualtour.aalto.fi

More about Aalto University:

Aalto.fiyoutube.com/user/aaltouniversitylinkedin.com/school/aalto-university/www.facebook.com/aaltouniversityinstagram.com/aaltouniversity

To view information about Workday Accessibility, please click here.

Please see more of our Open Positions here.

• Updated: 12.10.2026
• Published: 12.10.2026