Profitability of photovoltaic and battery systems on municipal buildings
Introduction
In countries in which grid parity has been reached for decentralized photovoltaic (PV) power plants, self-consumption is the main driver for the profitability of such systems. Stationary battery storage can increase the amount of self-consumed electricity significantly. Due to technological improvement and strong cost reductions of lithium-ion batteries, more and more PV systems are nowadays combined with batteries [1].
Current research focuses strongly on storage applications for residential buildings. Results from these studies demonstrate the impact of the load profile on the self-consumption rate (SCR), self-sufficiency rate (SSR) and the profitability of the PV battery system. Therefore, findings cannot directly be applied to other types of buildings and uses. With PV systems nowadays being installed on many schools, daycare centers, town halls and other community buildings, municipal properties are important applications of PV systems. Due to the good temporal match of demand and PV generation in these buildings, self-consumption rates are typically higher than for residential buildings. There are no studies, however, that analyze SCR, SSR and profitability of PV and battery storage systems on municipal properties specifically, and for different battery management strategies. The aim of this study is to close this gap and to provide realistic findings that can be used by project planners, municipal authorities and policy makers. For this purpose, the following research questions are formulated: What SCR and SSR can be realized for different kinds of municipal buildings and sizes of PV storage systems, and which variables influence SCR and SSR the most? Do the current framework conditions allow for an economic operation of PV storage systems on municipal properties? Which are the strongest influencing factors on the profitability?
Many studies have investigated self-consumption, self-sufficiency and profitability of PV battery storage systems, along with different charging and discharging strategies. The authors of [2] calculate SCR and SSR for different sizes of PV rooftop systems and battery storage, based on a reference household load profile. A model to estimate the internal rate of return for different PV plant and battery sizes, using simulated household load profiles, is developed by Ref. [3]. The authors of [4] determine the cost optimum constellation of PV system size and storage capacity, by varying both parameters. The authors of [5] calculate SSR, SCR and electricity cost for different PV system and battery sizes, based on the load profiles of two supermarkets. The authors of [6] derive equations to calculate SSR and SCR as a function of normalized PV system and storage size by applying an artificial neural network. They also investigate the impact of different discharging strategies on the grid and economic performance of the system. Their results are based on one residential load profile. The authors of [7] compare a forecast-based operation, based on linear optimization, to self-consumption maximizing operation for residential applications. According to Ref. [4], there are significant differences in SSR and SCR between simulated, aggregated and standard load profiles. The authors of [2] find that SSR varies by 23% for different households with equal (relative) PV system and storage size. The authors of [8] determine SCR and SSR for more than 2000 Swedish households and find differences in SCR of more than 20% for the same (relative) PV system and storage size. The authors of [9] focus on the profitability of PV in a municipal setting, but look at the aspect of communities of (mostly residential) buildings rather than municipal buildings. To the best knowledge of the authors, no studies specifically focusing on municipal buildings are yet available.
The conditions of prices, support schemes and solar irradiation for Germany are taken as the basis for the analysis in this study, but general findings can also be transferred to other locations with similar framework conditions. Linear programming is used to simulate two types of battery management systems (BMS), one that minimizes electricity procurement cost and one that additionally keeps the maximum power injection to the public grid as low as possible. The latter aspect is useful for PV support schemes that imply limits on the possible feed-in power. The model inputs are PV generation data for one year with a resolution of 15 min, system parameters of the PV plant, the battery and the inverter, and 15 min measured load profiles of 101 municipal properties. The model is run for 110 different PV storage system sizes for each property. SCR, SSR and losses that occur due to different feed-in power limits are calculated. Taking into account electricity prices, feed-in tariffs as well as acquisition and maintenance cost, the internal rate of return (IRR) and the break-even price for the battery at which the IRR would just be zero are calculated. The results are compared to each other, and explanatory variables for potential variances are identified using the generalized linear regression model (GLM) and analysis of variance (ANOVA).
The remainder of the paper is structured as follows: The applied model, assumptions made and the data input are presented in Sec. 2. In Sec. 3, results of both models, their statistical evaluation and a sensitivity analysis are presented. Sec. 4 discusses the main findings. Finally, Sec. 5 concludes the findings.
Section snippets
Method and data
The programming problem used for minimum cost battery management is presented in Sec. 2.1, and the parameter assumptions and data inputs are described in Sec. 2.2.
Results
The linprog function of Matlab was used to calculate all results for each of the 110 system configurations applied to the available consumption data sets. Results are presented for the self-consumption rate in Sec 3.1, for the self-sufficiency rate in Sec. 3.2, and for profitability in Sec. 3.3.
Discussion
All results presented in this study are based on data in 15 min time resolution. It must be noted that every time aggregation reduces the information on actual power peaks of shorter duration to some extent. It has been shown by previous studies that this leads to a systematic overestimation of self-consumption, and an underestimation of losses due to feed-in limits [18]. find that the self-sufficiency rate of a 1 kWp/MWh PV plant is increased by 1.5% if hourly data is used, compared to data in
Conclusions
This study investigated self-consumption rates, self-sufficiency rates and profitability measures for PV battery systems using the load data of 101 municipal properties in 15 min time resolution. Profitability for all system size configurations were evaluated through the measure of internal rate of return. In addition, the battery prices necessary to bring a PV battery system to a net present value of zero have been calculated. Two different algorithms were used, one that simulates battery
CRediT authorship contribution statement
Rafael Hirschburger: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software. Anke Weidlich: Methodology, Supervision, Validation, Visualization, Writing - review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (20)
- et al.
What drives the profitability of household pv investments, self-consumption and self-sufficiency?
Appl. Energy
(2017) - et al.
Optimization of self-consumption and techno-economic analysis of pv-battery systems in commercial applications
Appl. Energy
(2016) - et al.
Comparison of different discharge strategies of grid-connected residential pv systems with energy storage in perspective of optimal battery energy storage system sizing
Renew. Sustain. Energy Rev.
(2017) - et al.
Impacts of a forecast-based operation strategy for grid-connected pv storage systems on profitability and the energy system
Sol. Energy
(2017) - et al.
Solar photovoltaic-battery systems in Swedish households – self-consumption and self-sufficiency
Appl. Energy
(2016) - et al.
Profitability of pv sharing in energy communities: use cases for different settlement patterns
Energy
(2019) - et al.
Analysis of the maximal possible grid relief from pv-peak-power impacts by using storage systems for increased self-consumption
Appl. Energy
(2015) - et al.
Optimization of operational cost for a grid-supporting pv system with battery storage
Renew. Energy
(2016) Limitations for the feed-in power of residential photovoltaic systems in Germany – an overview of the regulatory framework
Sol. Energy
(2018)- et al.
Quantifying self-consumption linked to solar home battery systems: statistical analysis and economic assessment
Appl. Energy
(2016)
Cited by (10)
Techno-economic model of nonincentivized self consumption with residential PV systems in the context of Dominican Republic: A case study
2022, Energy for Sustainable DevelopmentSimulation or estimation?—Two approaches to calculate financial benefits of energy communities
2022, Journal of Cleaner ProductionCitation Excerpt :On a single-building level, installing PV systems is particularly interesting in the context of optimising self-consumption for office (Martín-Chivelet and Montero-Gómez, 2017) and commercial buildings (Merei et al., 2016). For these buildings, as well as municipal buildings, PV systems are of increased financial value due to a good match of supply and demand (Hirschburger and Weidlich, 2020). However, the sharing of PV system electricity, along with its profitability, is also increasingly evaluated for residential multi-apartment buildings (Fina et al., 2018, 2020b; Roberts et al., 2019).
The optimal design, synthesis and operation of polygeneration energy systems: Balancing life cycle environmental and economic priorities
2021, Energy Conversion and ManagementCitation Excerpt :The revenue from the electricity injected into the grid is constant (10.0 c€/kWh) during all the analysed periods. Furthermore, Table 3 summarises the specific investment costs (purchase and installation) of the selected sub-systems [6,55,83,84] and the maintenance costs [58], in line with the literature. The lifecycle assessment (LCA) method has been used to characterise the greenhouse gas emissions of the investigated polygeneration systems.