Dr. S.H. (Simon) Tindemans

Dr. S.H. (Simon) Tindemans

Profile

Biography

Simon Tindemans is an Assistant Professor in the Intelligent Electrical Power Grids group at TU Delft, focusing on uncertainty and risk management for power systems. With his team, he works at the intersection of electrical engineering, (computational) statistics and computer science. He has a background in theoretical (bio-)physics, but has since gained over 10 years' experience in the power engineering domain, contributing to the energy transition. 

Simon is an active member of the Reliability, Risk and Probability Applications subcommittee of the IEEE Power and Energy Society. He serves as the Secretary of the Resource Adequacy Working Group and the vice-chair and secretary of the Probability Applications of Common-Mode and Dependent Events Working Group. He is an advocate of open science (open access, open data and open source).

Simon teaches on subjects of power system reliability, machine learning, and power system resource planning, and has contributed to online courses on EdX (via DelftX).


Research: uncertainty and risk management

My team researches various aspects of uncertainty representation, risk assessment and planning for power systems. Three research streams are detailed below, with applications in:

  • Congestion management
  • Resource adequacy assessment
  • Operations aware planning

Swarm flexibility

Electric vehicles, smart appliances and responsive end users provide a significant potential for demand response. Even when the contribution of a single user is small, the collective contribution of all users can be very large. What is not clear is how this aggregate flexibility is best unlocked, so we investigate various building blocks and control strategies: 

  • Aggregate flexibility of EV charging and vehicle-to-grid (V2G)
  • Real-time market-based control (decentralised markets with probabilistic forecasts)
  • Residential load shaping with network tariffs
  • Control of heterogeneous storage aggregates
  • Safe reinforcement learning

Postdocs: Na Li, Amirreza Silani

PhD researchers: Hazem AbdelghanyRoman HennigNanda PandaQisong Yang

Risk assessment and mitigation

Reliability is a key design objective of the energy system, but quantifying the impact of decisions on system reliability is not straightforward, because reliability metrics are defined statistically. Moreover, because power systems are highly reliable, quantifying system reliability is effectively the study of rare high-impact low-probability (HILP) events, where bottlenecks in data availability and computation are especially challenging. On the interface between risk quantification and decision-making, We work on:

  • Efficient sampling-based methods for risk assessment, including importance sampling, active learning and multilevel Monte Carlo
  • Embedding operational risks in long-term system planning
  • Mathematical definition of risk metrics

Postdoc: Mojtaba Moradi-Sepahvand

PhD researcher: Ensieh Sharifnia

Data-driven modelling

In modern electrical power systems, sensors and other sources generate a growing stream of data. Efficient operation of the system requires extracting useful information and actionable insights from this data. Our work in this area focuses on:

  • Statistical learning of predictive models, including surrogate models of complex energy system simulators
  • Quantification of sampling uncertainty in real-world and simulated samples
  • Modelling multivariate dependencies, including applications to anomaly detection, data synthesis and data imputation

PhD researchers: Chenguang WangKutay Bölat

Former team members

PhD researchers (formal co-supervision)

MSc students  Theses available at  https://repository.tudelft.nl/islandora/search/?collection=education
  • Lotte Zwart (2022, with Stedin), "Congestion forecasting using a custom loss function"
  • Thomas Georgiou (2022, with Qirion/Alliander), "Interpretation of the Machine Learning Fault Classification Model Results Using Explainable Features"
  • Kevin Dankers (2022), "Profit optimized machine learning aided electricity storage; Comparison between privately owned and neighborhood shared"
  • Archana Ranganathan (2021, with Qirion/Alliander), "Automatic Identification of Fault Types in the Distribution Network using Supervised Learning"
  • Sai Suprabhath Nibhanupudi (2021, with Phase to Phase), "State Estimation in Medium Voltage Distribution Networks"
  • Ramon Mengerink (2021), "Cross-Border Participation in Capacity Mechanisms"
  • Devendra Kulkarni (2020, with Qirion/Alliander), "Unsupervised Learning to Locate Weak Spots in the Medium Voltage Grid"
  • Jules Zweekhorst (2020, with TenneT), "The development of a two day ahead power forecasting model for an offshore windpark"
  • Julian Betge (2020, with Alliander), "Monte Carlo Sampling Techniques for the Efficient Estimation of Risk Metrics of a Stochastic Distribution Grid Power Demand Model"
  • Subhitcha Ramkumar (2020), "Real Time Market Based Control of Flexible Distributed Energy Resources"
  • Medha Subramanian (2020, with TenneT), "Optimising Grid Topology Reconfiguration using Reinforcement Learning"
  • María Miranda Castillo (2019, with ENTSO-E), "Evaluation of missing capacity and resource adequacy in an interconnected power system"
  • Shreyas Nikte (2019), "Investigation of active learning techniques for dynamic Time-of-Use (dToU) tariff policy design for residential users"
  • Jennie Christiaanse (2019, with HyTEPS), "Algorithm for Determining the Hosting Capacity of Independent PV, or EV Charger Systems"
  • Jonathan Budez (2019), "Quantifying the Contribution of distributed flexible loads to congestion management"
  • Sotiris Dimitrakopoulos (2019, with DEPsys), "Linear state estimation method for distribution grids"
  • Roberto Francica (2019, with Alliander), "Assessment of machine learning algorithms for the purpose of heat pump detection based on load profiles and temperature readings"
  • Cheng-Kai Wang (2018), "Urban building energy modeling using a 3D city model and minimizing uncertainty through Bayesian inference"
  • Rob de Nie (2018, with DWA), "Detecting a change in building electricity consumption patterns from electricity master meter data"

Read more

/* */