Installation

This project uses a conda environment to manage the Python version and required packages. Follow below steps to start

1. Clone the repository

git clone https://github.com/ronitd07/Master-Thesis.git
cd Master-Thesis

2. Create conda environment

conda create --name myenv python=3.14

3. Activate the environment

conda activate myenv

4. Install the local project using pyproject.toml:

pip install .

5. Place the charline and charmap json files before running simulation

Navigate to your .tespy folder in your HOME directory HOME/.tespy. Create a folder named data, if it does not exist. In this folder, you can place the two json-files char_lines.json and char_maps.json for your characteristics which are under the folder charline and charmaps. Or follow below steps to generate the charline and charmaps using the scripts

6. To generate charlines and charmaps

To generate charlines and charmaps run the scripts

python src/models/create_charline.py
python src/models/create_charmap.py

7. Data preprocessing

The data preprocessing is done by running the below script

python src/models/data_cleaning.py

8. Running the Main Simulation Scripts

Simulation with characteristic lines

python src/models/MVV_GKM_simulation_charline.py

Simulation with characteristic maps

python src/models/MVV_GKM_simulation_charmap.py

Simulation with real compressor powers

python src/models/MVV_GKM_simulation_compressorP.py

8. Results validation

The results from the simulation are validated by running the below script. Make sure you have run the simulation in above step before running the validation scripts below or else the result files will be missing.

python src/models/Error_metrics.py

9. Frequency distribution plot

To plot the COP residual frequency distribution plot run below script

python src/plots/plotbar.py

10. COP residual per Speedline boxplot

To plot the COP residual per Speedline for the results with charmap simulation run below script

python src/plots/plot_COP_error_per_Speedline_boxplot.py

Input files

• Manheim_data_original.xlsx - Unprocessed data received
• Manheim_data_cleaned.xlsx - processed input data. Also after Column 30 it has the results from the real compressor simulation run like compressor powers, efficiency, igva, scaling factor
• Manheim_data_cleaned_automated.xlsx - processed input data after running the data_cleaning.py script but will not have the results from real compressor power runs.

Simulation and Calibration of Large-Scale Heat Pumps Using Measurement Data

This repository contains the Python code, data processing scripts, TESPy models, calibration routines, and plotting scripts used for my master's thesis:

Simulation and Calibration of Large-Scale Heat Pumps Using Measurement Data

The work focuses on the modelling and calibration of a large-scale heat pump system using measurement data. The model is implemented in Python using TESPy, CoolProp, pandas, NumPy, Matplotlib, and scikit-learn.

Project Overview

The main objective of this project is to simulate the offdesign operation of a large-scale heat pump and calibrate the model against measured plant data.

The project includes:

• TESPy-based heat pump cycle modelling
• Compressor characteristic map evaluation
• Offdesign simulation
• Calibration of compressor performance
• Comparison between simulated and measured COP
• Error analysis using metrics such as RMSE, MAPE, R², and residual plots
• Generation of figures and tables for thesis documentation

Repository Structure

ma_ronit/
│
├──charline                                     # stores the charline.json files used in simulation
├──charmaps                                     # stores the charmaps.json files used used in simulation
│
├── data/process_data
│   ├── Manheim_data_cleaned.xlsx               # Cleaned and preprocessed data
│   ├── Manheim_data_original.xlsx              # Original unprocessed measurement data
│   ├── Manheim_data_cleaned_automated.xlsx     # Cleaned and preprocessed data using automaed script [data_cleaning.py](src/models/data_cleaning.py)
│   ├── fits_for_maps.xlsx                      # Speedline data for fits
│   └── *.json                                  # Design json files
│
├── results/            
│   └── *.csv                                   # Simulation and calibration results
│
├── src/
│   ├── models/
│   │    ├── MVV_GKM_simulation_charmap.py       # Heat Pump simulation using char maps
│   │    ├── MVV_GKM_simulation_charline.py      # Heat Pump simulation using char lines
│   │    └── MVV_GKM_simulation_compressorP.py   # Heat Pump simulation using real compressor Powers
│   └── plots/                                   # Contains the plotting scripts used
│
├── pyproject.toml                               # Python project configuration
├── environment.yml                              # Conda environment file
└── README.md                                    # Project documentation