Andrea Scrocca

The growing participation of photovoltaic generation in electricity markets requires forecasts consistent with day-ahead bidding deadlines. These deadlines often prevent the use of the most recent production measurements as forecasting inputs, making realistic data-availability assumptions essential. This study develops a market-aligned forecasting framework combining deterministic prediction and uncertainty estimation for a medium-voltage microgrid hosting 1 MW of distributed photovoltaic capacity. Three models are compared: seasonal autoregressive integrated moving average with exogenous variables (SARIMAX), random forest, and long short-term (LSTM) networks. Uncertainty is evaluated using a historical baseline and four data-driven approaches: linear quantile regression based on random forest forecasts, quantile regression averaging, quantile regression forests, and conformalized quantile regression forests. Sensitivity analyses examine forecasting configurations and data availability. Random forest achieves the best deterministic performance. Direct forecasting of aggregated production yields an R² of 0.84 and a mean absolute error of 13.36 kW, with accuracy comparable to aggregated single-meter forecasts. Quantile regression forests perform best overall among probabilistic methods, with a mean pinball loss of 9.64 kW and 94% coverage for the nominal 90% prediction interval. Conformalization improves calibration overall but not consistently under challenging conditions. These results support uncertainty-aware day-ahead scheduling under realistic market constraints.

The increasing penetration of distributed energy resources and end-use electrification are making flexibility a key resource for distribution system operators. Local flexibility markets are emerging as a promising mechanism, but their economic attractiveness for balancing service providers remains uncertain. This paper quantifies the realistic economic value that an aggregation of distributed energy resources can obtain by participating in a local flexibility market, while explicitly accounting for the interaction with implicit flexibility, driven by electricity-bill optimization. The analysis focuses on PoliGrid, the multi-energy microgrid of Politecnico di Milano, participating in MiNDFlex, the local flexibility market developed by Unareti. A block-based mixed-integer linear programming framework is used to optimize PoliGrid operation over a full year. The scheduling stage determines the optimal upward power bands to be offered in the monthly auctions. The operation stage simulates real system performance through a rolling-horizon approach, considering forecast errors and service activation orders. The study compares configurations with and without a 2 MW/4 MWh BESS and evaluates alternative market design choices. Results show that, without storage, market participation is never economically convenient. Conversely, with a BESS, annual savings increase from about 43 k€/year under implicit flexibility alone to about 148 k€/year when implicit and explicit flexibility are combined. Sensitivity analyses highlight the importance of baseline selection, the operational challenges associated with capacity-limitation products, and the benefits of shorter auction horizons and more granular daily products. These findings highlight the role of storage in unlocking LFM revenues and the strong impact of market design on flexibility value.

The increasing penetration of distributed energy resources and end-use electrification are making flexibility a key resource for distribution system operators. Local flexibility markets are emerging as a promising mechanism, but their economic attractiveness for balancing service providers remains uncertain. This paper quantifies the realistic economic value that an aggregation of distributed energy resources can obtain by participating in a local flexibility market, while explicitly accounting for the interaction with implicit flexibility, driven by electricity-bill optimization. The analysis focuses on PoliGrid, the multi-energy microgrid of Politecnico di Milano, participating in MiNDFlex, the local flexibility market developed by Unareti. A block-based mixed-integer linear programming framework is used to optimize PoliGrid operation over a full year. The scheduling stage determines the optimal upward power bands to be offered in the monthly auctions. The operation stage simulates real system performance through a rolling-horizon approach, considering forecast errors and service activation orders. The study compares configurations with and without a 2 MW/4 MWh BESS and evaluates alternative market design choices. Results show that, without storage, market participation is never economically convenient. Conversely, with a BESS, annual savings increase from about 43 k€/year under implicit flexibility alone to about 148 k€/year when implicit and explicit flexibility are combined. Sensitivity analyses highlight the importance of baseline selection, the operational challenges associated with capacity-limitation products, and the benefits of shorter auction horizons and more granular daily products. These findings highlight the role of storage in unlocking LFM revenues and the strong impact of market design on flexibility value.

The increasing penetration of distributed energy resources and end-use electrification are making flexibility a key resource for distribution system operators. Local flexibility markets are emerging as a promising mechanism, but their economic attractiveness for balancing service providers remains uncertain. This paper quantifies the realistic economic value that an aggregation of distributed energy resources can obtain by participating in a local flexibility market, while explicitly accounting for the interaction with implicit flexibility, driven by electricity-bill optimization. The analysis focuses on PoliGrid, the multi-energy microgrid of Politecnico di Milano, participating in MiNDFlex, the local flexibility market developed by Unareti. A block-based mixed-integer linear programming framework is used to optimize PoliGrid operation over a full year. The scheduling stage determines the optimal upward power bands to be offered in the monthly auctions. The operation stage simulates real system performance through a rolling-horizon approach, considering forecast errors and service activation orders. The study compares configurations with and without a 2 MW/4 MWh BESS and evaluates alternative market design choices. Results show that, without storage, market participation is never economically convenient. Conversely, with a BESS, annual savings increase from about 43 k€/year under implicit flexibility alone to about 148 k€/year when implicit and explicit flexibility are combined. Sensitivity analyses highlight the importance of baseline selection, the operational challenges associated with capacity-limitation products, and the benefits of shorter auction horizons and more granular daily products. These findings highlight the role of storage in unlocking LFM revenues and the strong impact of market design on flexibility value.

In the context of urban decarbonization, multi-energy microgrids (MEMGs) are gaining increasing relevance due to their ability to enhance synergies across multiple energy vectors. This study presents a block-based MILP framework developed to optimize the operations of a real MEMG, with a particular focus on accurately modeling the structure of electricity and natural gas bills. The objective is to assess the added economic value of integrating a battery energy storage system (BESS) under the assumption it is employed to provide implicit flexibility—namely, bill management, energy arbitrage, and peak shaving. Results show that under assumed market conditions, tariff schemes, and BESS costs, none of the analyzed BESS configurations achieve a positive net present value. However, a 2 MW/4 MWh BESS yields a 3.8% reduction in annual operating costs compared to the base case without storage, driven by increased self-consumption (+2.8%), reduced thermal energy waste (–6.4%), and a substantial decrease in power-based electricity charges (–77.9%). The performed sensitivity analyses indicate that even with a significantly higher day-ahead market price spread, the BESS is not sufficiently incentivized to perform pure energy arbitrage and that the effectiveness of a time-of-use power-based tariff depends not only on the level of price differentiation but also on the BESS size. Overall, this study provides insights into the role of BESS in MEMGs and highlights the need for electricity bill designs that better reward the provision of implicit flexibility by storage systems.

Recent European regulations promote distributed energy resources as alternatives to centralized generation. This study compares utility-scale and distributed photovoltaic (PV) systems coupled with Battery Energy-Storage Systems (BESSs) in the Italian electricity market, analyzing different battery sizes. A multistage stochastic mixed-integer linear programming model, using Monte Carlo PV production scenarios, optimizes day-ahead and intra-day market offers while incorporating PV forecast updates. In real time, battery flexibility reduces imbalances. Here we show that, to ensure dispatchability—defined as keeping annual imbalances below 5% of PV output—a 1 MW PV system requires 220 kWh of storage for utility-scale and 50 kWh for distributed systems, increasing the levelized cost of electricity by +13.1% and +1.94%, respectively. Net present value is negative for BESSs performing imbalance netting only. Therefore, a multiple service strategy, including imbalance netting and energy arbitrage, is introduced. Performing arbitrage while keeping dispatchability reaches an economic optimum with a 1.7 MWh BESS for utility-scale systems and 1.1 MWh BESS for distributed systems. These results show lower PV firming costs than previous studies, and highlight that under a multiple-service strategy, better economic outcomes are obtained with larger storage capacities.

Recent European regulations promote distributed energy resources as alternatives to centralized generation. This study compares utility-scale and distributed photovoltaic (PV) systems coupled with battery energy storage systems (BESS) in the Italian electricity market, analyzing different battery sizes. A multistage stochastic MILP model, using Monte Carlo PV production scenarios, optimizes day-ahead and intraday market offers while incorporating forecast updates. In real time, battery flexibility reduces imbalances. Here we show that, to ensure dispatchability—defined as annual imbalances below 5% of PV output—a 1 MW PV system requires 200 kWh of storage for utility-scale and 100 kWh for distributed systems, increasing LCOE by 12.6% and 3.8% respectively. NPV is negative for BESS performing imbalance netting only. Therefore, a multiple service-strategy, including imbalance netting and energy arbitrage, is introduced. However, maintaining dispatchability while performing arbitrage reaches economic optimum with a 1.7 MWh BESS for utility-scale systems and 1.1 MWh BESS for distributed systems. These results show lower PV firming costs than previous studies, and highlight that under a multiple-service strategy, better economic outcomes are obtained with larger storage capacities.