Abhijith SivaprasadanM.Sc. Sustainable Energy Engineering · KTH · Stockholm, Sweden
Research / PhD / Portfolio
I’m interested in research-engineer and doctoral work across thermal-fluid and process modelling, dynamic energy-system simulation, integrated energy systems, uncertainty quantification and machine learning for physical systems. My work combines a published KTH master’s thesis with measurement-chain commissioning, dynamic thermal-storage modelling, industrial decarbonisation methodology and evidence-first power-system research. I frame research questions around identifiable assumptions, independently check the implementation where possible and report non-results and validation gaps without disguising them.
01 / Selected work
A focused selection. Open a case study for methods, results and limitations, or follow the skill dossiers below for the complete related record.
Industrial thesis at Siemens Energy Finspång combining preliminary NI-DAQ/LabVIEW commissioning and supervised heater-failure troubleshooting with a 15-run numerical CFD/CHT comparison of Pulsatorn reducer geometries.
ANSYS Fluent · CFD/CHT · k-omega SST · NI-DAQ · LabVIEW · Siemens NX
Public-data research freeze (23 October 2026) on Kerala's electricity-system stress, with observed SLDC daily targets, ERA5 retrospective demand analysis, hydro timing and bounded transmission screening. Not a validated planning model.
Python · PyPSA · OSeMOSYS · ERA5 · HiGHS · GIS · pytest
Techno-economic screening of industrial thermal energy storage with state-of-charge-dependent discharge constraints derived from dynamic models. Packed-bed, molten-salt and PCM cases are compared with explicit verification, sensitivity and validation boundaries.
Python · Pyomo · HiGHS · Modelica · FMI/FMPy · SALib
Physics-informed power-system dynamics surrogate with deterministic reference-simulator fallback. Frozen IEEE-14 mixed-workload tests report 1.57–1.95× full-route speedup; no operational safety claim.
PyTorch · ANDES · Python · SciPy · CI
Source-linked AI-assisted title/available-abstract screening of 538 OpenAlex heat-pump adoption candidates (144 include, 341 exclude, 53 uncertain). Independent 240-record human reference screening and blinded second review remain unfinished.
Python · scikit-learn · OpenAlex · pytest · GitHub Actions
An open-source two-boiler steam-and-power twin with a carbon-aware Pyomo/HiGHS scheduler, OpenModelica FMI 2.0 simulation and a local Streamlit advisory GUI. Synthetic and uncalibrated; not an ABB product or live control system.
Python · Modelica · Pyomo · HiGHS · FMI/FMPy · Streamlit · pytest
How do radiation approximations affect calculated heat transfer? Published-data fixtures, a discrete-ordinates solver and scientific regression tests. Original work and third-party source retain distinct provenance.
Python · NumPy · SciPy
Local electricity-investment GUI and shared Pyomo/HiGHS planner, with audited market coverage and 175 extracted public input references. Synthetic and uncalibrated; normalization, price-target decisions and weather conversion still gate historical fitting.
Python · Pyomo · HiGHS · Streamlit · NumPy · pytest
Modern Fortran educational gas-turbine and CCGT simulator. The v2.0.0-beta.1 source-only prerelease is archived on Zenodo (10.5281/zenodo.23679426); release CI passed, while GUI and industrial asset validation remain unaccepted.
Fortran 2008 · Python · Win32/GDI+
02 / Experience
Seven months embedded at Siemens Energy Finspang in the Fluid Dynamic Lab. Conducted numerical investigation of steady-state thermo-fluid performance of a reducer geometry for a high-temperature dynamic pressure sensor calibration rig (Pulsatorn). Built compressible CFD and conjugate heat transfer models in ANSYS Fluent (k-omega SST), conducted three-level mesh independence study, and applied Biot number analysis to decompose thermal performance. Identified flow-regime asymmetry between geometry variants (Ma 0.990 near-choked vs Ma 1.006 supersonic vena contracta). Also commissioned NI-DAQ measurement chains, modified LabVIEW VIs, commissioned the measurement chain for a planned validation campaign at up to 700 C, and conducted a formal root cause analysis of a heater failure resulting in an approved project scope revision. Used Siemens NX and Teamcenter PLM2020 throughout. Supervisor: Prof. Jens Fridh, KTH. Thesis: TRITA-ITM-EX 2026:14.
ANSYS Fluent · NI-DAQ · LabVIEW · Siemens NX · Teamcenter PLM2020 · Conjugate Heat Transfer · Compressible CFD · Root Cause Analysis
Contributed to KTH plastic pyrolysis research by reviewing reactor concepts and cost-analysis drivers for oil extraction from polymer waste, supporting early-stage technical and economic feasibility evaluation.
Developed and independently validated a quantitative methodology for industrial energy performance mapping, covering KPI/EnPI design, load-driver logic, metering-gap assessment, deviation detection from noisy operational data and regulatory comparison. The work was desk-based method development and decision-support framing rather than hands-on plant execution, with electrical utilities and compressed air treated as relevant electrification pathways. Proprietary site detail is not republished.
Python · ISO 50001 · EU EED · KPI/EnPI Design · Energy Performance Mapping · Industrial Energy Analysis · Structured Reporting
03 / Skills & evidence
Each skill opens its own page with all related public projects, roles, coursework, training and supporting material.
04 / Education
Completed the M.Sc. programme, including the 30 ECTS thesis within the 120 ECTS total; degree certificate in process. Specialisation: Energy Supply through Clean Conversion Technologies. Grade A in Numerical Heat Transfer and AI Applications. The record also includes a 3 ECTS Aalto/Unite! elective.
Mechanical engineering foundation with coursework, design competitions and the final-year interactive robot project.
Credited exchange elective covering lifecycle and circular-economy perspectives on energy storage; not a separate degree.
05 / Supporting material
Research interests describe future directions, not completed experiments or institutional endorsement. Nuclear energy is an emerging direction; it is not presented as an existing qualification. The thesis is a numerical investigation with measurement-chain commissioning; heater failure occurred before high-temperature testing could begin, so it does not establish experimental validation. Kerala2040 and TES Discharge Screen publish explicit screening, verification and validation boundaries; independent models are not operational forecasts.
Source case studies remain authoritative. Private inputs and restricted project details are not published.
Let’s connect
For doctoral applications and research collaborations.
abhijithsivaprasadan@gmail.com