BESS Simulation Tool

A Python-based simulation tool for Battery Energy Storage System (BESS) reserve provision, implementing the robust market-based management strategy described in:

Baltputnis, K., Broka, Z., Cingels, G., Sīlis, A., & Junghāns, G. (2024). Robust market-based battery energy storage management strategy for operation in European balancing markets. Journal of Energy Storage, 102, 114082. https://doi.org/10.1016/j.est.2024.114082

The input and output data concerning this paper is available on Zenodo: https://doi.org/10.5281/zenodo.18199324

The tool supports simultaneous FCR (Frequency Containment Reserve) and aFRR (automatic Frequency Restoration Reserve) market participation with market-based SOC restoration exclusively via the intraday market, fully compliant with EU System Operation Guideline requirements for Limited Energy Reservoirs (LERs).

Methodology overview

The simulation implements a worst-case activation anticipation approach where non-delivery of contracted reserves is never permitted. Key principles:

• No activation overfulfilment or FCR deadband utilization for SOC restoration
• SOC restoration exclusively via scheduled ID market transactions
• Full compliance with EU SOG requirements for LER FCR providers

Three decision-making nodules

1. BESS SOC Management (in simulation.py): Prepares ID market bids ensuring sufficient charge for worst-case reserve activation scenarios. Calculates worst-case energy requirements considering FCR, FRR capacity, scheduled ID trades, and self-discharge losses.

2. LER Management in Alert State (in bess_model.py): Implements reserve mode/normal mode transitions with 5-min aFRR full activation time for transitions. Tracks ΔTminLER criterion fulfillment (30 min full FCR activation during alert state) and manages recovery status post-alert (max 2 hours per SOG).

3. Voluntary FRR Energy Bid Preparation (in bess_model.py): Estimates additional FRR bids using surplus BESS capacity after ensuring worst-case scenario handling. Includes SOC violation subroutine to verify bids don't cause issues in other time periods.

SOC management strategies

• Conservative Strategy (for LER): Takes advantage of lenient criteria for LER FCR providers during alert state. Lower worst-case energy requirements, less active ID trading.

• Active Strategy (for non-LER): Requires continuous BESS availability irrespective of system state. More frequent ID trades but ensures full FCR provision without reserve mode transitions.

Market parameters (Continental Europe defaults)

• FCR: Symmetric capacity product, ±200 mHz full activation, ±10 mHz deadband
• aFRR: 5-min full activation time
• Market Time Unit (MTU): 15 minutes
• ID GCT: Configurable (e.g., 60 min for Baltic, 5 min for Germany)

Features

• Multi-BESS simulation: Two BESS units operating as a fleet with capacity allocation
• Reserve market participation: FCR and FRR capacity/energy market modeling
• LER support: Full implementation including reserve mode, ΔTminLER tracking, and recovery
• SOC management: Automatic ID market trading based on worst-case energy calculations
• Voluntary FRR bidding: Automatic bidding of available capacity with exhaustive worst-case validation
• Minimum cycling: Optional BESS-to-BESS energy exchange to meet cycling requirements
• Trade netting: Physical energy exchange lesser than sum of individual market deliveries
• Comprehensive output: Excel results and interactive matplotlib visualizations

Installation

Prerequisites

• Python 3.8 or higher
• pip package manager

Dependencies

pip install numpy pandas matplotlib openpyxl

Project structure

bess_sim/
├── main.py                  # Entry point - run this
├── constants.py             # Physical and market constants (MTU, time deltas)
├── utils.py                 # Utility functions (logging, rounding, ID clearing)
├── config.py                # Configuration management from settings.xlsx
├── data_input.py            # Frequency and FRR activation data loading
├── bess_model.py            # BESS class: energy calculations, LER procedures, voluntary FRR
├── simulation.py            # Main loop: SOC management, Alert State, reserve activation
├── output.py                # Excel results and matplotlib figures
├── settings.xlsx            # Simulation settings
├── output_template.xlsx     # Excel output template
├── frequency_*.csv          # Frequency input data
├── FRR_*.csv                # FRR activation input data
└── README.md                # This file

Configuration

Settings file (settings.xlsx)

1. BESS Technical Parameters: Power ratings, capacity, efficiency, SOC limits, losses
2. BESS Simulation Settings: Initial SOC, LER qualification, availability, SOC management strategy
3. Reserve Provision Settings: FCR/FRR capacity obligations (uniform or per-MTU)
4. Market Settings: GCT timings (ID, FRR), preparation times, frequency parameters (deadband, full activation deviation), ΔTminLER, max recovery time

Input files

• Frequency file: CSV/Excel with Time and Value columns (1-second to 1-minute resolution recommended)
• FRR activation file: CSV/Excel with Start, End, and MW columns (positive = up-regulation)

Usage

python main.py

The simulation resolution is configurable (1-minute recommended for balance of accuracy and performance).

Output

1. Results folder: Named {scenario_name} Run {n}/ containing:

   • Excel results with detailed time-series
   • Log file with console messages and warnings
   • Copy of settings used

2. Results Excel: Per-timestep data for both BESS units:

   • SOC trajectories
   • Contracted FCR/FRR availability
   • Activation expectations and actual delivery
   • Delivery failures (if any)
   • Reserve mode and recovery status

3. Interactive figures:

   • Row 1: SOC trajectory and ID schedule with Alert State highlighting
   • Row 2: FCR activation vs frequency deviation with Normal/Reserve mode indication
   • Row 3: FRR activation vs contracted capacity

Module overview

Module	Description
main.py	Entry point, orchestrates simulation flow
constants.py	Time constants (MTU=15min, hour, etc.)
config.py	Reads settings.xlsx, provides configuration dataclasses
data_input.py	Loads and resamples frequency/FRR data
bess_model.py	BESS class with E_worst calculations, FCR/FRR logic, LER reserve mode, voluntary FRR bidding
simulation.py	Main simulation loop, SOC management decisions, Alert State detection, energy delivery
output.py	Excel output writing, matplotlib figure generation
utils.py	Helper functions (custom_print, true_round, ID_clearing)

Validation

The strategy has been validated against:

• Extreme worst-case scenario: 6-hour continuous full FCR + aFRR activation
• Realistic scenarios: Full-year simulations with German (CE) and Finnish (Nordic) frequency data
• Sensitivity analysis: ID GCT varied from 15-105 minutes

Acknowledgments

This research is funded by the Latvian Council of Science, project "Multi-functional modelling tool for the significantly altering future electricity markets and their development (SignAture)", project No. lzp-2021/1-0227.

---