Load flow and 24-hour quasi-static time-series simulation of a radial distribution network with integrated PV generation, Battery Energy Storage (BESS), and EV charging, implemented in Python using pandapower. Results are compared against PSS/E — all 33 buses agree within 0.0001 pu.
| Metric | pandapower | PSS/E v33 |
|---|---|---|
| Min bus voltage (pu) | 0.9131 (Bus 18) | 0.9131 (Bus 18) |
| Max bus voltage (pu) | 1.0000 (Bus 1) | 1.0000 (Bus 1) |
| Buses below 0.95 pu | 21 | 21 |
| Total active power loss (MW) | 0.2027 | — |
| Max difference between tools | — | 0.0001 pu ✓ |
Comparison result: pandapower and PSS/E produce identical load flow results for the IEEE 33-bus system. Maximum voltage difference across all 33 buses is 0.0001 pu — within numerical precision of both Newton-Raphson solvers.
| Metric | Base Case | With DER |
|---|---|---|
| Min bus voltage (pu) | 0.9131 (Bus 18) | 0.9405 (Bus 29) |
| Buses below 0.95 pu | 21 | 6 |
| Voltage improvement | — | +2.74 % |
| Reduction in violations | — | 71 % |
Test system: IEEE 33-bus radial distribution network (Baran & Wu) Base voltage: 12.66 kV Total base load: 3.715 MW + j2.300 MVAr
| Asset | Bus | Rating | Basis for bus selection |
|---|---|---|---|
| PV Unit 1 | 14 | 2.0 MW peak | Lowest voltage in base case |
| PV Unit 2 | 31 | 2.0 MW peak | Second weakest bus |
| BESS | 31 | 0.5 MW / 2.0 MWh | Co-located with PV for loss reduction |
| EV Charging | 28 | 1.5 MW peak | Mid-feeder representative node |
BESS rule-based dispatch strategy :
- Charge when PV output > 30 % of rated and SOC < 90 % (hours 06:00–15:00)
- Discharge during evening demand peak and SOC > 20 % (hours 18:00–22:00)
- Efficiency: 95 %
The base case load flow was independently solved in two tools:
| Tool | Type | Solver | Language |
|---|---|---|---|
| pandapower | Newton-Raphson | Python 3 | |
| PSS/E | Newton-Raphson | -- |
Both tools used identical network data:
- Same bus topology (IEEE 33-bus)
- Same impedance values (per unit on 100 MVA, 12.66 kV base)
- Same load data (32 constant-power loads)
- Same slack bus (Bus 1)
Result: All 33 buses match within 0.0001 pu — confirming that the pandapower model is correctly implemented and suitable for DER integration studies.
der_load_flow_IEEE33bus/
├── src/
│ ├── main.py # runs full simulation
│ ├── network.py # Network builder and DER asset creation
│ ├── simulation.py # Load flow, BESS dispatch, time-series loop
│ ├── plots.py # All visualisation functions
│ └── compare_psse_pandapower.py # PSS/E vs pandapower comparisonn
├── results/ # Generated plots and CSV
│ ├── 01_voltage_profile_comparison.png
│ ├── 02_timeseries_voltage.png
│ ├── 03_der_dispatch_and_soc.png
│ ├── 04_power_losses.png
│ ├── 05_voltage_violations.png
│ ├── 06_comparison.png
│ └── timeseries_results.csv
├── psse/
│ └── IEEE33bus.sav # PSS/E v33 saved case file
├── requirements.txt
└── README.md
# 1. Install dependencies
pip install -r requirements.txt
# 2. Run the full DER simulation (generates plots 01-05 and CSV)
python src/main.py
# 3. Run the PSS/E cross-validation comparison (generates plot 06)
python src/compare_psse_pandapower.pyAll plots are saved to results/ automatically.
This project is part of my broader research on DER-integrated microgrids, which includes MPC-based power management and hybrid energy storage systems. Related publications:
- Jena, C.J., Ray, P.K. — Power Quality Enhancement and Power Management of PV-HESS Based Grid-Tied Microgrid Using Model Predictive Control, IEEE Transactions on Industry Applications, 2024.
- Jena, C.J., Ray, P.K. — Power Allocation Scheme for Grid-Interactive Microgrid with Hybrid Energy Storage System Using Model Predictive Control, Journal of Energy Storage, 2024.
- Jena, C.J., Ray, P.K. — Power Management in Three-Phase Grid-Integrated PV System with Hybrid Energy Storage System, Energies (MDPI), 2023.
Python · pandapower · NumPy · pandas · matplotlib · PSS/E
Baran, M.E. and Wu, F.F. (1989) Network reconfiguration in distribution systems for loss reduction and load balancing, IEEE Transactions on Power Delivery, 4(2), pp. 1401–1407.
MIT