A main challenge for integrating the intermittent photovoltaic (PV) power generation remains the accuracy of day-ahead forecasts and the establishment of robust performing methods. The purpose of this work is to address these technological challenges by evaluating the day-ahead PV production forecasting performance of different machine learning models under different supervised learning regimes and minimal input features. Specifically, the day-ahead forecasting capability of Bayesian neural network (BNN), support vector regression (SVR), and regression tree (RT) models was investigated by employing the same dataset for training and performance verification, thus enabling a valid comparison. The training regime analysis demonstrated that the performance of the investigated models was strongly dependent on the timeframe of the train set, training data sequence, and application of irradiance condition filters. Furthermore, accurate results were obtained utilizing only the measured power output and other calculated parameters for training. Consequently, useful information is provided for establishing a robust day-ahead forecasting methodology that utilizes calculated input parameters and an optimal supervised learning approach. Finally, the obtained results demonstrated that the optimally constructed BNN outperformed all other machine learning models achieving forecasting accuracies lower than 5%.

No abstract

Distributed generation (DG) systems are growing in number, diversifying in driving technologies and providing substantial energy quantities in covering the energy needs of the interconnected system in an optimal way. This evolution of technologies is a response to the needs of the energy transition to a low carbon economy. A nanogrid is dependent on local resources through appropriate DG, confined within the boundaries of an energy domain not exceeding 100 kW of power. It can be a single building that is equipped with a local electricity generation to fulfil the building’s load consumption requirements, it is electrically interconnected with the external power system and it can optionally be equipped with a storage system. It is, however, mandatory that a nanogrid is equipped with a controller for optimisation of the production/consumption curves. This study presents design consideretions for nanogrids and the design of a nanogrid system consisting of a 40 kWp photovoltaic (PV) system and a 50 kWh battery energy storage system (BESS) managed via a central converter able to perform demand-side management (DSM). The implementation of the nanogrid aims at reducing the CO2 footprint of the confined domain and increase its self-sufficiency.

No abstract

No abstract