Methodological Analysis of Cellular Energy Systems under Increasing Renewable Energy Integration

Keywords: Renewable energy, cellular-energy system, electric charging, supply security, energy-atlas

The increasing integration of renewable energy sources is fundamentally transforming urban and regional energy systems. Centralized, unidirectional supply structures are being replaced by decentralised systems with multidirectional energy flows, necessitating new organisational and analytical approaches to ensure supply security. Cellular energy systems have emerged as a promising concept, structuring infrastructures into interconnected, semi-autonomous units that integrate generation, storage, distribution, consumption, and control. This paper presents a modular, data-driven framework for analysing such systems, incorporating building-level characteristics, socio-demographic trends, and stochastic technology diffusion. The approach is demonstrated using the city of Wittichenau, Germany, as a case study [1], generating high-resolution energy demand profiles up to 2050 to assess sector-coupled interactions and infrastructure loads.

Introduction

The expansion of renewable energy technologies is fundamentally transforming energy supply systems from centralised, large-scale structures with unidirectional flows to decentralised systems characterised by distributed resources and multidirectional energy flows, further intensified by the increasing integration of local energy conversion units. Traditional sector-specific planning is increasingly inadequate for ensuring supply security and system stability in decentralized, interconnected energy systems. Instead, new models—such as cellular energy systems—are required, which explicitly account for bidirectional flows and interactions between energy carriers. These systems consist of interconnected energy cells, each functioning as a semi-autonomous unit that integrates generation, conversion, distribution, storage, consumption, and control, enabling local optimization while remaining coordinated within the overall system. [2]

The structure of cellular energy systems, which is shown in Figure 1, is characterized by a high degree of interconnection between different technologies for energy generation and conversion. Electrical energy, thermal energy (heat and cold), and gaseous energy carriers (e.g., methane and hydrogen) must be considered in an integrated manner. Sector coupling significantly increases system complexity, particularly when transforming existing infrastructures with closely linked electrical and thermal demands. Within the project “Cellular Energy Systems for the Transformation of the Energy Supply in the Suburban Region,” the team at Dresden University of Technology developed a comprehensive, spatially explicit methodology for analysing coupled energy systems [1]. Applied to the model city of Wittichenau (eastern Saxony, Germany), the approach integrates energy-related factors with socio-demographic and economic developments in its scenario design.

Figure 1. Structure of a “Cellular Energy System”. [2]

Methodological framework

The proposed methodology follows a modular and iterative approach (shown in Figure 2) and consists of three main components:

·         building-related modules,

·         technology diffusion modules, and

·         high-resolution temporal energy demand modelling.

Figure 2. Methodological workflow for the analysis of a cellular energy system. Building-related and technology modules are evaluated sequentially within an annual iterative simulation framework. Final system states of each year serve as input values for the subsequent simulation step.

The first step involves analysing the existing building stock with respect to typologies, thermal properties, and socio-demographic developments (component i). Vacancy rates and renovation dynamics are estimated based on population forecasts and stochastic allocation methods. Renovation measures are prioritized according to buildings’ specific heat demand using a gamma distribution, whereby buildings with higher heat demand have a greater probability of selection and are thus preferentially renovated. For renovated buildings, specific heat demand values are assigned using the heat atlas methodology described in [3], ensuring realistic energy performance characteristics [3, 4].

Building-related modules provide the basis for subsequent technology modules (component ii), in which energy technologies—such as photovoltaic systems, heat pumps, electrical storage, and EV charging infrastructure—are assigned to individual buildings. The modular simulation structure allows for flexible extension, e.g., to incorporate wind energy expansion. Technology allocation is performed stochastically based on weighted criteria, including exclusion constraints, building suitability, and externally defined expansion targets. In addition, the timing of system renewal is considered: for the base year (2023), heat generator ages in newer buildings (constructed from 2009 onward) are derived from building age, while for older buildings, age distributions are statistically assigned using scaled BDEW data for eastern Germany [5]. Heating technologies are parameterized according to the boundary conditions specified in Table 1.

Table 1. Calculated service life of heat generators according to VDI 2067. [6].

The simulation is modular to ensure flexibility and transferability to different regions. Each module is evaluated annually at the building level, with actions determined by weighted stochastic processes, rendering the model non-deterministic. Consequently, the results are twofold: building-specific scenario pathways and, through repeated simulations, statistically robust insights into aggregate indicators such as total heat demand or average specific consumption. This also enables the identification of recurring spatial trends, e.g., vacancy patterns. Based on this framework, high-resolution building-specific thermal and electrical load profiles (component iii) are generated for selected reference years. These profiles support the assessment of infrastructure requirements across electricity, gas, and district heating networks, including interactions with integrated technologies. Crucially, system evaluation is driven by time-resolved power rather than aggregated energy, necessitating dynamic analysis. Thermal systems can be dimensioned based on peak loads, while electrical infrastructure—particularly low-voltage grids—requires high-resolution load profiles for accurate assessment.

Results

Building and Technology Modules

The thermal energy atlas [6] forms the foundation of the methodological framework, enabling a building-level characterization of heat demand. While such spatial detail exceeds typical municipal planning needs, it allows flexible aggregation; for Wittichenau, full resolution is retained to ensure transparency. Buildings are classified using the TABULA typology, complemented by geospatial data (e.g., footprints, 3D city models, OpenStreetMap) and an extended scheme distinguishing key building types. Seven energy performance classes are defined based on German building regulations. Due to limited data availability, renovation status is inferred from external features (e.g., façade, insulation, windows, roof), ensuring practical applicability. Based on these attributes, building-specific heat demand values are derived. Simulations indicate a consistent population decline across all scenarios, leading to rising vacancy rates and a substantial reduction in energy demand. These effects are incorporated into renovation dynamics up to 2050, with building-level distributions shown for selected years (e.g., 2025, 2035, 2050) – see Figure 3.

Figure 3. Building-specific distribution of the simulation results for the years 2025, 2035 and 2050 for one simulation run. [1]

Renovation activity, assumed at 1% annually in Saxony [8], is modelled stochastically using weighted selection. In line with the EU Energy Performance of Buildings Directive (EPBD) [9], buildings with the highest specific heat demand are prioritized. Following renovation, heat demand values are updated using the heat atlas methodology to ensure realistic performance.

Technology modules extend the building-level results by assigning adoption probabilities based on exclusion criteria, technical suitability, and external constraints. For photovoltaics, factors such as roof orientation, usable area, and local regulations—e.g., restrictions in Wittichenau’s historic center—are considered. The results for PV are illustrative shown in Figure 4, which indicate that by 2050, a substantial share of buildings is equipped with PV systems and heat pumps.

Figure 4. Exemplary PV system expansion in Wittichenau for the years 2025, 2035, and 2050 for a simulation run. [1]

Similar methodological principles are applied to the modules for heat pumps, electrical energy storage and electromobility. In the electromobility module, a distinction is made between public and private charging infrastructure, as well as between coordinated and uncoordinated charging strategies.

Time series determination

Building and technology modules provide annual thermal and electrical energy demands, but system design requires time-resolved power profiles. Therefore, analysis is conducted dynamically across both sectors. In the thermal sector, peak loads determine system sizing, while in the electrical sector, high-resolution profiles are essential for low-voltage grid assessment. Annual demands are thus converted into time-resolved load profiles. Thermal profiles are derived top-down using degree days based on local weather data, assuming an indoor temperature of 20 °C and building-specific heating limits; the heating period excludes summer conditions. Daily loads are calculated from annual demand, with domestic hot water assumed constant. These are then disaggregated to 15-minute intervals using temperature-dependent or stochastic load profiles, while hot water demand is assigned stochastically. Resulting profiles are illustrated in Figure 5.

Figure 5. Exemplary annual load profile of heat demand with τ = 24h (left) and selected daily load profiles τ = 15min (right) for a single-family home based on this.

These load profiles form the basis for analysing heat generators (e.g., heat pumps) as interfaces to the electrical grid, directly linking thermal and electrical sectors. Electrical load time series are derived using a bottom-up approach (Figure 6), based on an algorithm according to [7] that explicitly models all building-level electrical consumers. Electrical load profiles are derived from a statistical estimate of building occupancy, combined with typical high-resolution usage profiles for device groups (e.g., household appliances, IT, and consumer electronics). Aggregation of these loads yields the total building demand, supplemented by additional loads such as heat pumps based on thermal profiles. This framework also enables flexible modelling of EV charging. Two scenarios are considered [1]:

·         uncoordinated charging, allowing unrestricted, independent charging, and

·         building-level coordinated charging, limiting peak load.

The impacts on load profiles are illustrated in Figure 7.

Figure 6. Bottom-up approach for determining the load time series for house connections. [1,7]

Figure 7. Scenarios for charging using the example of an apartment building (house connection) (left: coordinated/ right: uncoordinated charging). [1]

Figure 7 shows that in uncoordinated charging operation, very high electrical load peaks of up to P = 100 kW per house connection can occur, which is significantly higher than the currently available maximum connection power of P = 50 kW. In coordinated charging, on the other hand, no power levels above P = 50 kW occur. At the same time, the average charging time increases to around τ = 1000 h at lower charging power, while in uncoordinated operation charging times of up to τ = 200 h are achieved.

Conclusion

Urban and suburban energy systems are rapidly transforming due to increasing renewable integration and decentralisation. The presented methodology offers a comprehensive framework for analysing such systems within the concept of cellular energy systems by combining building-level analysis, stochastic technology allocation, and high-resolution temporal modelling. A key strength is the integrated treatment of electricity, heat, and other energy carriers alongside socio-demographic and economic factors. Implemented as an open-source tool, the approach is transferable to other regions and supports energy system analysis and planning in the energy transition.

Nomenclature

Acknowledgements

Further analyses can be found on the website of the underlying research project https://zellsys.de/en/.

The project on which this article is based was funded by the Federal Ministry of Economics and Climate Protection and the Federal Ministry of Economics and Export Control under the funding code 46SKD186X.

References

[1]     Grun, P.; Seidel, P.; Seifert, J.; Kreutziger, M.; Potyka, Schegner, P.; ; M.; Keller, J.: Zellsys – Zellulare Energieversorgung zur Umgestaltung der Energieversorgung im suburbanne Raum, Forschungsbericht, TU Dresden, 04/2026, VDE Verlag ISBN: 978-3-8007-6700-7.

[2]     Seifert, J.; Schegner, P.: Zellulare Energiesysteme – Grundlagen, Teilsysteme, Märkte, Rahmenbedingungen, Praxisbeispiele; ISBN 978-3-8007-5557-8, VDE Verlag 2023.

[3]     Grun, P.; Seidel, P.; Seifert, J.: Gebäudescharfe Analyse der Wärmebedarfssituation im suburbanen Raum, HLH BD.75 (2024), Nr.03.

[4]     Grun, P.: Analyse eines zellularen Energiesystems im suburbanen Raum am Beispiel der Stadt Wittichenau (südlich von Hoyerswerda), Diplomarbeit, TU Dresden 2023.

[5]     BDEW; Studie – Wie heizt Deutschland, 2019,  https://www.bdew.de/media/documents/20190812_BDEW-Studie-Wie-heizt-Deutschland.pdf, Zugriff: 04/2025.

[6]     VDI (2012): VDI 2067 Blatt 1 – Wirtschaftlichkeit gebäudetechnischer Anlagen. Grundlagen. Düsseldorf: VDI.

[7]     Dickert, J.: Synthese von Zeitreihen elektrischer Lasten basierend auf technischen und sozialen Kennzahlen – Grundlage für Planung, Betrieb und Simulation von aktiven Verteilungsnetzen, Dissertation TU Dresden 2015.

[8]     Sächsisches Staatsministerium für Umwelt und Landwirtschaft Sächsisches Staatsministerium, Hrsg.: Aktionsplan Klima und Energie des Freistaates Sachsen. URL: https://www.klima.sachsen.de/download/SMUL-Aktionplan_final_web.pdf (visited on 05. 03. 2023).

[9]     Europäische Kommission, Hrsg.: The Energy Performance of Buildings Directive (EPBD). URL: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=OJ:L_202401275 (visited on 01.12. 2024).