Impact of Retail Tariff Choice and Virtual Wallet on Forecast-Based PV Sharing in Residential Energy Communities
J. Serrano González, J.T. Villalonga Palou, J.M. Riquelme Santos, J.M Roldán Fernández 1.Department of Electrical Engineering, University of Seville, Seville, Spain 2026-06-27

Abstract

Collective self-consumption schemes typically require dynamic sharing coefficients—hourly allocation ratios per customer—to be submitted during the month preceding settlement, making reliable month-ahead, hour-by-hour demand forecasts essential. This paper presents a practical framework that generates these forecasts from historical consumption data and derives compliant allocation coefficients, while explicitly assessing how the economic value of forecasting depends on the retail tariff and compensation rules. Several lightweight persistence-based predictors are used to capture recent individual load behavior, and their outputs are combined through an ensemble that learns data-driven weights from past errors. The framework is evaluated on real hourly data from 18 residential customers under three representative tariff settings: PVPC, a market-based contract without virtual wallet, and a market-based contract with virtual wallet. Results show that the ensemble consistently improves upon individual predictors and achieves annual bills close to an oracle benchmark, while revealing that tariff design—especially whether export credits can offset fixed monthly charges—materially affects both total costs and the sensitivity of outcomes to PV size and export price.

Key words: collective self-consumption, demand forecasting, electricity tariffs, ensemble learning.

References

[1]             R. Luthander, J. Widén, D. Nilsson, and J. Palm, “Photovoltaic self-consumption in buildings: A review,” Appl. Energy, vol. 142, pp. 80–94, Mar. 2015, doi: 10.1016/j.apenergy.2014.12.028.

[2]             F. D. Minuto and A. Lanzini, “Energy-sharing mechanisms for energy community members under different asset ownership schemes and user demand profiles,” Renew. Sustain. Energy Rev., vol. 168, p. 112859, Oct. 2022, doi: 10.1016/j.rser.2022.112859.

[3]             I. F. G. Reis, I. Gonçalves, M. A. R. Lopes, and C. H. Antunes, “Collective self-consumption in multi-tenancy buildings–To what extent do consumers’ goals influence the energy system’s performance?,” Sustain. Cities Soc., vol. 80, p. 103688, May 2022, doi: 10.1016/j.scs.2022.103688.

[4]             N. Li and Ö. Okur, “Economic analysis of energy communities: Investment options and cost allocation,” Appl. Energy, vol. 336, p. 120706, Apr. 2023, doi: 10.1016/j.apenergy.2023.120706.

[5]             A. Kortetmäki, J. Ylipaino, J. Koskela, K. Kallioharju, and P. Järventausta, “The impact of metering methods on collective self-consumption: Insights from multi-dwelling buildings in Finland,” Energy Build., vol. 319, p. 114511, Sep. 2024, doi: 10.1016/j.enbuild.2024.114511.

[6]             E. Llera-Sastresa, J. Á. Gimeno, J. L. Osorio-Tejada, and P. Portillo-Tarragona, “Effect of Sharing Schemes on the Collective Energy Self-Consumption Feasibility,” Energies, vol. 16, no. 18, p. 6564, Jan. 2023, doi: 10.3390/en16186564.

[7]             J. T. Villalonga Palou, J. Serrano González, and J. M. Riquelme Santos, “Estimation of Energy Distribution Coefficients in Collective Self-Consumption Using Meta-Heuristic Optimization Techniques,” Sustainability, vol. 16, no. 7, p. 2741, Jan. 2024, doi: 10.3390/su16072741.

[8]             J. T. Villalonga Palou, J. Serrano González, J. M. Riquelme Santos, and J. M. Roldán Fernández, “A novel weight-based ensemble method for emerging energy players: an application to electric vehicle load prediction,” Energy AI, vol. 20, p. 100510, May 2025, doi: 10.1016/j.egyai.2025.100510.