4-year PhD Funded Position in Greenhouse Climate Control & Learning Theory at TU Delft, Netherlands

Recent trends in Controlled Environment Agriculture (CEA) development, design, and operations related to energy saving and greenhouse gas emission reduction, are mainly focused on the ventilation process, which typically considers the use of window and mechanical air treatment based on average climate measurements. Airflow affects crop transpiration, growth, development, yield and quality, but despite its importance, related control strategies in practice are often very crude and rule-based without incorporating any complex plant/microclimate interactions or economic considerations.

The state-of-the-art approaches in optimal climate control of greenhouses are based on implementing economic objective functions exploiting a time scale decomposition between short-term climate control/energy use, and long-term crop management goals. While several algorithms have shown promising results in energy savings and crop yield, most of these methods have only been tested in simulation, and make use of average climate measurements, which are then used to control the overall climate setpoints. Awareness of micro-climate insight and fine-grained, model-based control of locally applied ventilation is lacking in all these approaches.

Vacancy Overview

• Institution: Delft University of Technology (TU Delft)
• Faculty: Faculty of Mechanical Engineering
• Project Name: GreenControl Consortium
• Position: PhD Candidate (Systems & Control / Applied Mathematics)
• Contract: 1.0 FTE (38 hours/week); 4-year structured employment (1.5 years initial + 2.5 years extension upon positive evaluation)
• Salary Range: €3,059 to €3,881 gross per month (CLA Dutch Universities)
• Location: Delft, Netherlands
• Application Deadline: September 30, 2026

The Faculty of Mechanical Engineering at TU Delft is recruiting a full-time PhD candidate to advance data-driven control theory and machine learning for Controlled Environment Agriculture (CEA).

The doctoral position is embedded in GreenControl, an interdisciplinary research project uniting academic teams at TU Delft, Wageningen University, University of Twente, and TU Eindhoven, alongside commercial greenhouse developers, sensing technology providers, and plant breeders.

The GreenControl initiative aims to shift greenhouse operations from indirect, average-climate rule-based management toward direct, crop-centric microclimate control. The target is to cut total energy consumption by 25% and reduce energy costs by 35% while sustaining optimal photosynthetic activity and crop yield.

Research Objectives and Methodology

Traditional optimal climate control algorithms operate on averaged spatial measurements and macro-level setpoints, overlooking localized microclimates and complex plant-air interactions. This PhD candidate will address these gaps by constructing control-oriented, reduced-order mathematical models using data collected from micro-sensors, plant imaging, and high-fidelity Computational Fluid Dynamics (CFD) simulations.

Key Technical Directions:

• Hybrid Data-Driven & Model Predictive Control: Formulate predictive control architectures using Koopman operator formalisms, reproducing kernel Hilbert spaces (RKHS), neural networks, and Sparse Identification of Nonlinear Dynamics (SINDy) to manage high-dimensional, nonlinear partial differential equations (PDEs).
• Airflow & Microclimate Regulation: Model localized airflow dynamics, CO2 delivery to leaf surfaces, temperature, and humidity using forced convection, tightly coupling crop biological targets with climate regulation.
• Energy Cost Optimization: Design real-time control policies that adapt lighting schedules, artificial CO2 dosing, active ventilation, and thermal screens based on real-time electricity price fluctuations and biological demand.
• Sensor & Actuator Placement: Optimize physical spatial configurations for sensor networks and fan/ventilation actuators to maximize benefit-to-cost ratios.
• Experimental Validation: Implement and iteratively validate control algorithms on high-fidelity computational simulations and physical greenhouse demonstrator setups.

Faculty and Graduate School Environment

The candidate will join the Faculty of Mechanical Engineering at TU Delft and enroll in the TU Delft Graduate School. The Graduate School provides structured doctoral education covering technical discipline-specific skills, research ethics, and transferable career competencies.

The candidate will work under the supervision of Prof. dr. ir. Tamas Keviczky within a multidisciplinary environment connecting control theory, fluid dynamics, and biological systems modeling.

Qualification Requirements

Applicants should possess a strong theoretical background in quantitative engineering disciplines paired with an interest in bio-physical system applications.

Required Qualifications:

• Master of Science (MSc) degree in systems and control, applied mathematics, mechanical engineering, computational engineering, or a closely related field.
• Background or interest at the intersection of systems theory, machine learning, PDEs, and biological/physical system dynamics.
• Written and oral English proficiency meeting TU Delft Graduate School entry standards.
• Experience in conducting or managing physical/biological system experiments is advantageous but not mandatory.

Employment Conditions and Benefits

This 4-year doctoral contract follows the Collective Labour Agreement (CLA) for Dutch Universities.

Benefit Component	Details
Gross Monthly Salary	€3,059 (Year 1) increasing to €3,881 (Year 4) for 38 hours/week
Contract Structure	1.5-year initial contract; progress evaluation at 15 months leads to a 2.5-year extension
Annual Allowances	8% Holiday allowance + 8.3% End-of-year bonus
Relocation Support	Coming to Delft Service assists with housing, settling in, and partner career support
Secondary Benefits	Customisable compensation package, healthcare discounts, monthly work costs contribution

Application Instructions

Submit your complete application through the TU Delft online portal by September 30, 2026. Emailed submissions are not accepted.

Required Application Documents:

1. A detailed Curriculum Vitae.
2. A 1-page motivation letter addressing your interest in this specific PhD topic.
3. Transcripts for both Bachelor’s and Master’s degrees (including coursework grades).
4. One or two written research samples (e.g., MSc thesis draft or peer-reviewed publication).
5. Contact details for two academic referees (requested if selected for an interview).
6. APPLY NOW

Direct scientific questions regarding the role to Prof. dr. ir. Tamas Keviczky (t.keviczky@tudelft.nl).

Note: As part of national knowledge security regulations, TU Delft conducts a standard risk assessment during the final recruitment stages to prevent unauthorized technology or knowledge transfer.

Frequently Asked Questions

What is the core target of the GreenControl project?

The GreenControl consortium aims to transition controlled-environment agriculture to direct crop-centric microclimate control, achieving a 25% energy reduction and a 35% drop in energy costs through microclimate sensing and automated control algorithms.

What mathematical methods will this PhD project focus on?

The project focuses on learning theory for PDEs, Koopman operator formalisms, SINDy (Sparse Identification of Nonlinear Dynamics), neural networks, function spaces (RKHS), and data-driven model predictive control (MPC).

How is the 4-year PhD contract structured at TU Delft?

Candidates receive an initial 1.5-year employment contract. Following a formal go/no-go assessment around month 15, the contract is extended for the remaining 2.5 years (4 years total).

What extra support is available for international candidates relocating to Delft?

TU Delft provides relocation assistance through the Coming to Delft Service, covering administrative guidance, networking events, and a Dual Career Programme to support accompanying partners in their job search in the Netherlands.