MiGUEL is continously optimized in terms of handling and outputs.

MiGUEL is a python-based, open-source simulation tool to design, simulate and evaluate the performance of photovoltaic-diesel-hybrid systems. MiGUEL is based on a matlab tool developed at the Technische Hochschule Köln (TH Köln). In the course of the research project Energy-Self-Sufficiency for Health Facilities in Ghana (EnerSHelF) the matlab tool was transferred to python, revised and additional components were added.
MiGUEL aims to provide an easy-to-use simulation tool with low entry barriers and comprehensible results. Only a basic knowledge of the programming language is needed to use the tool. For the system design, simulation and evaluation, only a small number of parameters is needed. The simulation can run without data sets provided by the user. The results are provided in the form of csv files for each simulation step and in the form of an automatically generated pdf report. The csv files are understood as raw data for further processing. The pdf report serves as a project brochure. Here, the results are presented clearly and graphically, and an economic and ecological evaluation of the system is carried out.

• Authors and contributors
• Content and structure
  • Main
  • Environment
  • Operator
  • Evaluation
  • Output
• Graphical user interface
• Database
• Project partners
• Dependencies
• References
• Appendix

The main author is Paul Bohn (@pdb-94). Co-author of the project is Silvan Rummeny (@srummeny) who created the first approach within his PhD. Other contributors are Moritz End (@moend95). Further assistance was provided by Sascha Birk (@pyosch). The development of the tool was supervised by Prof. Dr. Schneiders (TH Köln CIRE).

The basic structure of MiGUEL is displayed below.

The class Environment represents the energy system. It takes basic parameters such as time frame, location, economic and ecologic parameters. System components can be added to the Environment. The Operator runs the simulation and evaluation of the designed energy system. The class Report creates the pdf-report. The program is run by the main file.

The main file is used to run the program. The main file is the only time the user has to interact with the source code. The Environment, Operator and Report are created by the user.

The class Environment represents the energy system.

To create an instance of the class, the following parameters have to be provided. The list displays all input parameters, a brief description and the data type.

MiGUEL features the following system components. Each component can be added to the Environment by using a different function. The list displays the system components and the functions to add the components to the Environment.

The system component load represents the load profile of the subject under review. The load profile can be generated in two different ways.

1. Reference load profiles: In the course of EnerSHelF standard load profiles for Ghanaian hospitals were created. This daily standard load profile is implemented in the program. Since May 2023 the reference load profiles from the Bundesverband der Energie- und Wasserwirtschaft (BDEW) have been included. The reference load profiles are used in the german dispatch to simulate certain inistitutions. [17] To create a load profile from the reference load profiles, the annual electricity consumption needs to be returned to the function (annual_consumption). The reference load profiles have a 15min-time resolution.
2. Input via csv-file: If actual measurement data from the subject is available, the data can be returned to the program as a csv-file (load_profile). The csv file must contain two columns with the titles 'time' & 'P [W]'. ',' or ';' are used as separators; for decimal separation '.' or ',' are used depending on the setting.

The accuracy of the simulation results increases with the quality of the input data. Using the adjusted standard load profile will provide less accurate results compared to measured data. The library Load Profile Creator can be used to create load profiles based on the electric inventory of the subject.

If the resolution of the load profile does not match the environment time resolution, the resolution of the load profile will be adjusted by summarizing or filling in the values. If no annual load profile is provided, the load profile will be repeated to create an annual load profile.

The class Photovoltaic is based on the library pvlib [1]. There are three methods implemented to create PV systems:

1. Adding basic system parameters: Simplest way to create PV system with only basic parameters such as nominal power, surface tilt and azimuth, module and inverter power range. The class Photovoltaic will randomly choose a PV module, number of modules and an inverter that matches the parameters.
2. Selecting your modules and inverter: All system parameters such as module, number of modules, inverter, strings per inverter, modules per string, surface tilt and azimuth, ... need to be returned to the function. The modules and inverters featured in pvlib are stored in the MiGUEL database.
3. Provide measured PV data: Input of measured PV as a csv-file

pvlib will run the PV simulation based on the selected system parameters. The weather data for the project location is retrieved by the Environment. The data source is PVGIS hosted by the European Commission.

The class WindTurbine is based on the library windpowerlib [2]. To add wind turbines to the Environment the turbine type and the turbine height [m] need to be returned. The wind turbines featured in windpowerlib are stored in the MiGUELdatabase.

The weather data for the project location is retrieved by the Environment. The data source is PVGIS hosted by the European Commission. Inside the class WindTurbine the weather data is processed, so it can be used for the simulation.

The class grid represents the power grid. The power grid provides electricity to the energy system. Depending on the input of blackout data, a stable or unstable power grid is simulated. The possibility of feed-in is determined in the Environment. The grid is automatically added to the Environment if the parameter grid_connection is set to True.

The class DieselGenerator is based on a simplified, self created generator model. The model assumes that in the future generators with low-load capability are used in PV-diesel hybrid systems. In comparison to conventional diesel generators, low-load diesel generators are more fuel efficient and therefore reduce CO2-emissions [3]. The input parameters for diesel generators are displayed in the table below.

The fuel consumption for the generator is calculated every time step using the following equation. The equation was derived using characteristic values of a 150 kW diesel generator at loads of 0%, 25%, 50%, 75% and 100% [4].

fc(l) = - 1.66360855 x l ⁴ +3.96330272 x l ³ -3.19877674 x l ²+1.8990825 x l +0

fc = relative fuel consumption [%]   l = relative load [%]

The class Storage represents energy storage systems. The energy storage is represented by a basic model. The input parameters for storage systems are displayed in the table below:

The energy storage can be either charged or discharged at any time step. The following boundary conditions apply to loading and unloading. The memory can only be discharged to the minimum state of charge and charged to the maximum state of charge. The maximum charging or discharging power corresponds to the nominal power multiplied by the respective efficiency.

The simulation process is divided in three steps.

The system design is the only time the user needs to interact with the program code. Here the Environment (create Environment) and the system components are created (system components). The annual simulation and the system evaluation are carried out by the Operator.

The energy system type depends on the input parameters and the system components in the energy system. A distinction is made between off-grid systems and on-grid systems. On-grid systems are further divided into stable systems (without blackouts) and unstable systems (with blackouts). Depending on the type of energy system, different dispatch strategies are applied for the annual simulation.

RE = Renewable energies   ES = Energy storage   DG = Diesel generator

The figure displays the dispatch strategies for all system components. If a system component is not added to the system, this component will be skipped in the dispatch.

The two key parameters for the system evaluation are the Levelized Cost of Energy (LCOE) in US$/kWh and the CO2-emissions [t] over the system lifetime. The class Evaluation takes the Envrionemnet and the Operator as input parameters.

Note: The specific values for investment, operating and maintenance costs have been partially converted from euros to US$ (27.03.2023). The costs may differ depending on the exchange rate.

The LCOE are calculated according to Michael Papapetrou et. al. for every energy supply component [5]. The system LCOE is composed of the individual LCOEs of the system components, which are scaled according to the energetic share. The LCOE are calculated over the whole system lifetime. The LCOE includes the initial investment costs and the operation and maintenance costs. Costs for recycling are neglected in this evaluation. The investment and operation and maintenance cost are based on specific costs from literature values. The specific costs are scaled by the power (energy supply components) or capacity (energy storage).

The CO2-emissions are evaluated over the system lifetime. Included are the CO2-emissions during the production of the system component and the CO2-emissions emitted during the usage.

MiGUEL provides two types of outputs. The first output is a csv-file with every simulation time step. The csv-files can be used for further research or in depth analysis of the system behaviour. The csv-files do not include the system evaluation. The second output is the pdf-report. The report includes the most important results. The results are displayed graphically and will be explained briefly.

The csv-files display the raw data of the annual simulation. The file lists every time step of the simulation, the load and all system components, as well as their generation power.

The pdf-Report is automatically created by MiGUEL. It gives an overview of the simulation results and features the system evaluation based on the LCOE and CO2-emissions. The report is structured in the following chapters:

1. Introduction: Brief description of MiGUEL and EnerSHelF
2. Summary: Summary of the most important simulation results and system evaluation
3. Base data: Displays input parameters
4. Climate data: Solar and wind data from PVGIS at the selected location
5. Energy consumption: Load profile
6. System configuration: Overview of selected system components
7. Dispatch: Annual simulation results
8. Evaluation: System evaluation based on LCOE and CO2-emissions over system lifetime

The report focuses not only on the energetic results of the system evaluation but also on economic and ecologic parameters. This makes the results more comprehensible compared to the csv-files. The pdf-report can be used as a project brochure.

End of June 2023 a graphical user interface (GUI) has been implemented into MiGUEL to increase the usability of the tool. With the implemtation the entry hurdle is lowered even more. The GUI follows the logical process as described above. The following list gives an overview of the different tabs and a short description of their function:

1. Get started: Welcome Screen including a brief overview of MiGUEL and EnerSHelF. Select csv file format
2. Energy system: Input mask to create Environment class.
3. Weather data: Displays weather data from PVGIS at selected location.
4. Load profile: Input mask to add load profile to Environment.
5. PV system: Input mask to add PV systems to Environment.
6. Wind turbine: Input mask to add wind turbines to Environment.
7. Diesel Generator: Input mask to addd diesel generator to Environment.
8. Energy storage: Input mask to add energy storage to Environment.
9. Dispatch: Overview of system components. Runs dispatch and system evaluation.
10. Evaluation: Overview of system evaluation parameters. Creates outputs.

MiGUEL features a SQLite database in the directory /data/miguel.db. The following tables are included in the database:

For a full list of all dependencies see requirements.txt. This file will ask the user to install the dependencies automtically.

pandas

numpy

matplotlib

folium

geopy

fpdf

pvlib

windpowerlib

selenium

plotly

lcoe

global-land-mask

PyQT5

geonames

geopandas

lcoe

plotly

[1] William F. Holmgren, Clifford W. Hansen, and Mark A. Mikofski. “pvlib python: a python package for modeling solar energy systems.” Journal of Open Source Software, 3(29), 884, (2018). https://doi.org/10.21105/joss.00884

[2] Sabine Haas, Uwe Krien, Birgit Schachler, Stickler Bot, kyri-petrou, Velibor Zeli, Kumar Shivam, & Stephen Bosch. (2021). wind-python/windpowerlib: Silent Improvements (v0.2.1). Zenodo. https://doi.org/10.5281/zenodo.4591809

[3] PV Magazine; "Low-load generators make photovoltaic diesel applications cleaner and more efficient"; 06. October 2015; online available: Niedrig-Last-Generatoren machen Photovoltaik-Diesel-Anwendungen sauberer und effizienter

[4] Generator Source, LLC 1999-2023; Approximate Diesel Fuel Consumption Chart; online available: https://www.generatorsource.com/Diesel_Fuel_Consumption.aspx

[5] Michael Papapetrou, George Kosmadakis, Chapter 9 - Resource, environmental, and economic aspects of SGHE, Editor(s): Alessandro Tamburini, Andrea Cipollina, Giorgio Micale, In Woodhead Publishing Series in Energy, Salinity Gradient Heat Engines, Woodhead Publishing, 2022, Pages 319-353, ISBN 9780081028476, https://doi.org/10.1016/B978-0-08-102847-6.00006-1

[6] Vartiainen, E, Masson, G, Breyer, C, Moser, D, Román Medina, E. Impact of weighted average cost of capital, capital expenditure, and other parameters on future utility-scale PV levelised cost of electricity. Prog Photovolt Res Appl. 2020; 28: 439– 453. https://doi.org/10.1002/pip.3189

[7] Bjarne Steffen, Martin Beuse, Paul Tautorat, Tobias S. Schmidt, Experience Curves for Operations and Maintenance Costs of Renewable Energy Technologies, Joule, Volume 4, Issue 2, 2020, Pages 359-375, ISSN 2542-4351, https://www.sciencedirect.com/science/article/pii/S2542435119305793

[8] Lucas Sens, Ulf Neuling, Martin Kaltschmitt, Capital expenditure and levelized cost of electricity of photovoltaic plants and wind turbines – Development by 2050, Renewable Energy, Volume 185, 2022, Pages 525-537, ISSN 0960-1481, https://www.sciencedirect.com/science/article/pii/S0960148121017626

[9] Tyler Stehly, Philipp Beiter, and Patrick Duffy, National Renewable Energy Laboratory, 2019 Cost of Wind Energy Review, 2019, https://www.nrel.gov/docs/fy21osti/78471.pdf9

[10] James Hamilton, Michael Negnevitsky, Xiaolin Wang, The potential of variable speed diesel application in increasing renewable energy source penetration, Energy Procedia, Volume 160, 2019, Pages 558-565, ISSN 1876-6102, https://doi.org/10.1016/j.egypro.2019.02.206

[11] The EU Global Technical Assistance Facility for Sustainable Energy (EU GTAF), Sustainable Energy Handbook Module 6.1 Simplified Financial Models

[12] National Renewable Energy Laboratory, Utility-Scale Battery Storage, 2023, https://atb.nrel.gov/electricity/2022/utility-scale_battery_storage

[13] Fraunhofer ISE, Photovoltaics and Climate Change, 2020, https://www.ise.fraunhofer.de/content/dam/ise/de/documents/publications/studies/ISE-Sustainable-PV-Manufacturing-in-Europe.pdf

[14] Ozoemena, M., Cheung, W.M. & Hasan, R. Comparative LCA of technology improvement opportunities for a 1.5-MW wind turbine in the context of an onshore wind farm. Clean Techn Environ Policy 20, 173–190 (2018). https://doi.org/10.1007/s10098-017-1466-2

[15] Friso Klemann, University Utrecht, The environmental impact of cycling 1,600 MWh electricity - A Life Cycle Assessment of a lithium-ion battery from Greener Power Solutions (P. 35)

[16] Hao, H.; Mu, Z.; Jiang, S.; Liu, Z.; Zhao, F. GHG Emissions from the Production of Lithium-Ion Batteries for Electric Vehicles in China. Sustainability 2017, 9, 504. https://doi.org/10.3390/su9040504

[17] BBDEW Bundesverband der Energie- und Wasserwirtschaft e.V.; Standardlastprofile Strom; https://www.bdew.de/energie/standardlastprofile-strom/; 01.01.2017