• Energy Research

Abstract Storage technologies such as the battery energy storage system are expected to play a critical role in micro-grids. However, current simulation tools underestimate their operating costs, which jeopardises their efficient use and deployment. This paper proposes a strategy based on artificial intelligence and time series prediction for the planning and real-time management of a battery energy storage system in a micro-grid acting as a virtual power plant. An economic analysis was performed, and the findings suggested that the state of charge should be managed to avoid economic losses due to cycle ageing. In addition, the battery should be sized correctly to ensure its economic viability, which indicates that it should be optimised according to the load profile of the micro-grid. The proposed management strategy was demonstrated to avoid the economic losses observed with a non-managed storage system. In addition, ensuring that the micro-grid did not deviate from the load profile agreed upon with the transmission system operator was found to increase economic returns, reduce battery degradation and increase self-consumption with the co-benefit of reducing micro-grid fluctuations.

The increasing attention to environmental issues is forcing the implementation of novel energy models based on renewable sources, fundamentally changing the configuration of energy management and introducing new criticalities that are only partly understood. In particular, renewable energies introduce fluctuations causing an increased request of conventional energy sources oriented to balance energy requests on short notices. In order to develop an effective usage of low-carbon sources, such fluctuations must be understood and tamed. In this paper we present a microscopic model for the description and the forecast of short time fluctuations related to renewable sources and to their effects on the electricity market. To account for the inter-dependencies among the energy market and the physical power dispatch network, we use a statistical mechanics approach to sample stochastic perturbations on the power system and an agent based approach for the prediction of the market players behavior. Our model is a data-driven; it builds on one day ahead real market transactions to train agents behaviour and allows to infer the market share of different energy sources. We benchmark our approach on the Italian market finding a good accordance with real data.

AbstractThis paper presents the main features and the expected performance of the pilot solar power plant under construction in Ottana (Sardinia-Italy). The facility is based on a 600 kWe concentrating solar power (CSP) plant with thermal energy storage, and a 400 kWe concentrating photovoltaic (CPV) plant with electrochemical storage. The CSP plant uses linear Fresnel collectors, thermal oil as heat transfer fluid, a two-tank direct storage system and an ORC module. The CPV plant consists of 37 dual-axis trackers integrated with Sodium-Nickel batteries. The facility is characterised by the integration of different concentrating solar and storage technologies. The pilot power plant has been designed in order to produce electricity with scheduled profiles according to weather forecast.

The design of an Inner-Rotor Large-Airgap Permanent Magnet Synchronous Machine (IRLA-PMSM) is presented in this paper, whose distinctive feature consists of integrating a thick carbon-fibre cylindrical layer that acts as both a sleeve and a flywheel. This results in large airgap and makes simplified mechanical modelling generally used for sleeve design no more suitable. Therefore, an advanced mechanical modelling is presented in the paper, which is combined with an electromagnetic and energy modelling in order to make the analytical design procedure compliant with all the design targets and constraints. The proposed IRLA-PMSM is validated through finite element analyses, which regard both mechanical and electromagnetic aspects at different operating conditions.

A novel management strategy for Electric Vehicles (EVs) storage systems is proposed in this paper. It aims to enhance the Renewable Energy Sources (RES) exploitation hour by hour, but prioritizing the EV mobility requirements. A mathematical modelling of the mobility system is firstly developed in order to estimate, hour by hour, the amount of EVs available to exchange energy with the electric grid, together with the average state of charge of their corresponding batteries. Then, a management strategy is properly introduced in order to increase the RES hourly energy production as much as possible, providing ancillary services too. The proposed management system is properly validated by means of a simulation study referred to a weakly interconnected power system, which is characterized by a strong RES penetration level too. The simulation results highlight the effectiveness of the proposed management strategy and the potentiality of the EVs storage systems in enhancing the RES exploitation.

This paper deals with electromagnetic loss analysis and minimization in an integrated Flywheel Energy Storage System (FESS). The FESS consists of a large-airgap Surface-Mounted Permanent Magnet Synchronous Machine (SPM), whose inner rotor integrates a carbon-fiber flywheel, leading to a compact and efficient FESS. Electromagnetic losses minimization is achieved through an accurate SPM modelling, which accounts for both Joule and core losses, as well as for flux-weakening operation. Consequently, SPM maximum efficiency is obtained by injecting a proper demagnetization current component at each operating condition, which guarantee the optimal trade-off among the different losses’ contributions. The effectiveness of the proposed approach is validated through numerical simulations.

This paper presents an online energy management tool that suggests the most suitable size of a hybrid photovoltaic-battery energy storage system (PV-BESS) to residential prosumers based on their self-sufficiency expectations. An offline analysis of electricity generation and consumption expected from 128 residential prosumers has been carried out at first in order to find out their self-sufficiency map with different sizes of PV and BESS; this is carried out by the genetic algorithm based energy management (GA) presented in a previous work. Subsequently, a number of clusters have been defined, each of which groups prosumers that share similar self-efficiency maps; particularly, clustering has been carried out and refined by identifying the most significant features of prosumers belonging to the same cluster, as well as those that differentiate prosumers belonging to different clusters. As a result, it has been revealed that the habit of usage of some appliances, such as Heat Ventilation Air Conditioning system (HVAC) and water heater, and peak electricity consumption represent the most important features influencing clustering. Based on these outcomes, the proposed online energy management tool is able to assign a prosumer to the most suitable cluster just based on the answers to a few simple questions related to energy consumption habits, providing the corresponding self-efficiency map almost immediately. The results achieved by the proposed tool, which is currently running online, are promising and show that significant self-sufficiency increases can be obtained, allowing the proper choice of PV-BESS depending on specific prosumer's needs and expectations.