News

News from Saeed Salehi, Linköping University: papers, talks, positions and appointments.

2026

PhD recruitment under way

September 2026

The application period for the PhD position in the project Enabling flexible hydropower operation by AI-based active flow control closed on 31 August 2026, and the assessment and recruitment process has now started.

We received a large number of applications, so the process will take some time. Thank you to all applicants for your patience. The position is expected to start in autumn 2026.

Read more about the research on the project page.

Review paper on PINNs for the lid-driven cavity problem

August 2026

Our paper State-of-the-art implementations of PINNs for the lid-driven cavity problem – a critical review with future perspectives is published in Results in Engineering (vol. 32, 112399).

Authors: M. Sheikholeslami, S. Salehi, W. Mao, A. Eslamdoost and H. Nilsson. The work is part of Mohammad Sheikholeslami’s PhD project on multi-fidelity physics-informed neural networks at Chalmers University of Technology, which I co-supervise. See also the PINN research theme.

Read the paper (DOI: 10.1016/j.rineng.2026.112399)

Talk and session chair at the 21st OpenFOAM Workshop

July 2026

At the 21st OpenFOAM Workshop (OFW21) in Guimarães, Portugal, I presented Closed-loop control of vortex rope in a swirl generator using deep reinforcement learning and chaired the Multiphase Flows I session.

A swirl generator is a simplified model of the flow in a hydraulic turbine draft tube. For background, see the flow control research theme.

Section Editor of the OpenFOAM Journal

June 2026

I have joined the editorial team of the OpenFOAM Journal as Section Editor. I also serve on the OpenFOAM Turbomachinery Technical Committee within the OpenFOAM Governance.

New PhD position in AI-based flow control for hydropower

June 2026

A PhD position was advertised in the project Enabling flexible hydropower operation by AI-based active flow control, funded by an ÅForsk Foundation Early-Career Research Grant with co-funding from Linköping University. The project couples deep reinforcement learning with high-fidelity CFD to suppress harmful flow instabilities in hydraulic turbines.

The student will be based at Linköping University, with Håkan Nilsson (Chalmers University of Technology) as co-supervisor. The application deadline was 31 August 2026.

About the project

Invited talk in Lund

May 2026

I gave the invited talk Towards efficient control of turbulence at a meeting of the Swedish Society for Industrial Flow and Heat Transfer in Lund.

Talk and session chair at the ERCOFTAC Workshop on Machine Learning for Fluid Dynamics

March 2026

At the 3rd ERCOFTAC Workshop on Machine Learning for Fluid Dynamics at CWI in Amsterdam, I presented Transfer learning strategies for accelerating reinforcement-learning-based flow control and chaired the Reinforcement Learning (RL3) session.

The talk is based on the arXiv preprint of the same title.

Docent in Fluid Mechanics

February 2026

I have been awarded the title of Docent in Fluid Mechanics at Linköping University. The title recognises independent scientific and pedagogical competence and supports principal supervision of doctoral students.

Invited talk at LKAB

February 2026

I gave the invited talk Engineering applications of computational and data-driven fluid dynamics for LKAB, Kiruna (held online).

2025

Review paper on the lifetime of hydro turbines

December 2025

Our review Lifetime analysis of hydro turbines with focus on fatigue damage in a renewable energy system is published in Renewable and Sustainable Energy Reviews (vol. 228, 116578).

Authors: M. Nobilo, S. Salehi and H. Nilsson. The paper reviews how transient and off-design operation, increasingly common as hydropower balances wind and solar power, affects fatigue damage and lifetime of hydro turbines. It is part of Martina Nobilo’s doctoral work at Chalmers University of Technology, which I co-supervised. See also the flexible hydropower research theme.

Read the paper (DOI: 10.1016/j.rser.2025.116578)

Preprint on transfer learning for reinforcement-learning-based flow control

October 2025

The preprint Transfer learning strategies for accelerating reinforcement-learning-based flow control is available on arXiv (2510.16016). See the flow control research theme for background.

Joined Linköping University

August 2025

I joined Linköping University as Associate Professor of Fluid Mechanics in the Division of Applied Thermodynamics and Fluid Mechanics, Department of Management and Engineering. Before that I was a researcher at Chalmers Industriteknik and a postdoctoral researcher at Chalmers University of Technology.