KTH MJ2505 · Power flow · Grid planning · pandapower

Distribution Grid Study with EV, PV and Storage Impacts

Distribution-grid simulation study using a CIGRE MV network, 24-hour load profiles, EV charging profiles, transformer-level connection assumptions and N-1 contingency analysis.

GitHub repository

Evidence dashboard

Grid flexibility framed as physical constraint analysis.

Technical question

How do EV charging, PV generation and storage assumptions alter voltage profiles, line loading and N-1 security in a medium-voltage distribution network?

24 hload profiles N-1contingency 348EV profiles

Engineering contribution

Ran a base CIGRE MV power-flow study and exported voltage and line-loading results.
Used 24-hour load-shape data to move from static power flow to time-series grid behaviour.
Integrated EV charging profiles through additional low-voltage buses and transformer connections.
Compared contingency outputs to identify where flexibility measures need to target thermal loading rather than only voltage support.

Five-step methodology

Step 1 – Initial network state: base power flow analysis and N-1 contingency analysis on the CIGRE MV network to establish voltage and loading baselines; exports confirmed 16 bus rows with voltage and thermal margins.
Step 2 – Time-series simulation: 24-hour load-shape profiles added; time-series power flow run to analyse daily demand variation and its impact on line and transformer loading across all network buses.
Step 3 – EV integration: estimated number of EV charging stations based on penetration scenarios; new low-voltage buses and transformer connections added; load flow and N-1 contingency repeated to locate voltage and thermal bottlenecks introduced by EV charging. Results: line loading reached 110.6%, transformer loading reached 104.4% in the stressed case.
Step 4 – Load management and optimisation: load profiles optimised to minimise operational costs; PV systems and storage added to both base and EV-stressed configurations; optimised dispatch evaluated against the unmanaged baseline to quantify the benefit of flexibility.
Step 5 – Network losses minimisation: operational optimisation to minimise network losses through strategic EV charging scheduling; evaluated potential loss reductions across the full 24-hour horizon using the optimised profiles.

Relevance

Why this matters

This project supports Energy Systems Modelling because it connects decarbonisation technologies to the physical network constraints they create. EV charging, PV and storage are not just scenario labels; they change line currents, transformer loading and operational margins in a real distribution-grid calculation workflow.

Base case16-row voltage and line-loading exports 110.6%line loading under EV stress (N-1 case) 104.4%transformer loading under EV stress

The structured five-step approach isolates each intervention — adding EVs, then adding storage and PV, then switching from cost to loss minimisation — so the effect of each design choice is traceable rather than mixed into a single scenario comparison.