A fully automated transferable predictive approach was developed to predict photovoltaic (PV) power output for a forecasting horizon of 24 h. The prediction of PV power output was made with the help of a long short-term memory machine learning algorithm. The main challenge of the approach was using (1) publicly available weather reports without solar irradiance values and (2) measured PV power without any technical information about the PV system. Using this input data, the developed model can predict the power output of the investigated PV systems with adequate accuracy. The lowest seasonal mean absolute scaled error of the prediction was reached by maximum size of the training set. Transferability of the developed approach was proven by making predictions of the PV power for warm and cold periods and for two different PV systems located in Oldenburg and Munich, Germany. The PV power prediction made with publicly available weather data was compared to the predictions made with fee-based solar irradiance data. The usage of the solar irradiance data led to more accurate predictions even with a much smaller training set. Although the model with publicly available weather data needed greater training sets, it could still make adequate predictions.

The high share of power generation based on fluctuating renewable energy sources, especially wind and solar, has increased the levels of variability and uncertainty in power systems. The aim of this study is to develop a method for quantifying the variability of photovoltaic (PV) systems. The developed method investigates the power measurements of a PV system and quantifies its power and energy fluctuations in three steps. The first includes a classification of days into three classes according to the variability and power output of the PV system. The second consists of computing the empirical cumulative distribution functions of PV power ramps for the given classes. The third calculates the PV daily energy fluctuations based on predicted PV power output. This method of PV variability quantification was then applied to seven rooftop PV systems in different locations that feature different installed capacities, years of installation, orientations and solar cell types. The testing of this method indicates that it can be used to quantify the power and energy fluctuations of different PV systems, independent of their locations and technical characteristics. The proposed method was developed as part of a general approach for quantifying the flexibility potential of buildings and city districts with PV systems.

Although they are primarily installed for specific applications, decentralised energy systems, storage systems, and controllable loads can provide flexibility. However, this varies over time. This study investigates the fundamentals of flexibility provision, including quantification, aggregation, simulation, and impact on energy systems and the power grid. We extended our methods by integrating adjustments to calculate the flexibility potential of heat pumps (HPs) and heat storage (HS) systems, as well as by incorporating variability and uncertainty. The simulations revealed the relevance of energy systems operation to flexibility, e.g., 2 K deviation in HS temperature increased theoretical coverage by 16 percentage points. The results also proved that aggregating multiple systems could obviously enhance their flexibility potential, e.g., six investigated battery storage (BS) systems could have covered up to 20 percentage points more external flexibility requests than any individual unit. The provision of flexibility by decentralised energy systems can lead to energy surpluses or deficits. Such imbalances could have been fully balanced in a system- and grid-oriented manner in 44% of BS simulations and in 32% of HP-HS ones. Overall, the findings highlight the importance of the system- and grid-oriented operation of decentralised energy systems, alongside local optimisation, for a future energy infrastructure.