Photovoltaic generation has arisen as a solution for the present energy challenge. However, power obtained through solar technologies has a strong correlation with certain meteorological variables such as solar irradiation, wind speed or ambient temperature. As a consequence, small changes in these variables can produce unexpected deviations in energy production. Although many research articles have been published in the last few years proposing different models for predicting these parameters, the vast majority of them do not consider spatiotemporal parameters. Hence, this paper presents a new solar irradiation forecaster which combines the advantages of machine learning and the optimisation of both spatial and temporal parameters in order to predict solar irradiation 10 min ahead. A validation step demonstrated that the deviation between the actual and forecasted solar irradiation was lower than 4% in 82.95% of the examined days. With regard to the error metrics, the root mean square error was 50.80 W/m(2), an improvement of 11.27% compared with the persistence model, which was used as a benchmark. The results indicate that the developed forecaster can be integrated into photovoltaic generators' to predict their output power, thus promoting their inclusion in the main power network. (C) 2021 Elsevier Ltd. All rights reserved.

Energy distribution companies have experienced an increase in catastrophic failures in power systems due to the theft of grounding cable. The objective of this study is to develop an innovative technology which is able to detect the absence of grounding cable in medium and high voltage systems. The wireless technology developed communicates periodically with the system's central control unit to ensure reliable and safe operation of the installation. Because it is not economically profitable to modify power systems' infrastructure, this novel sensor is non-intrusive as well as self-supplying. Based on the validation step's results, we conclude that the sensor is ready to be tested in smart grid equipment to monitor the state of grounding cables. (C) 2020 Elsevier B.V. All rights reserved.

Electrical load forecasting plays a crucial role in the proper scheduling and operation of power systems. To ensure the stability of the electrical network, it is necessary to balance energy generation and demand. Hence, different very short-term load forecast technologies are being designed to improve the efficiency of current control strategies. This paper proposes a new forecaster based on artificial intelligence, specifically on a recurrent neural network topology, trained with a Levenberg-Marquardt learning algorithm. Moreover, a sensitivity analysis was performed for determining the optimal input vector, structure and the optimal database length. In this case, the developed tool provides information about the energy demand for the next 15 min. The accuracy of the forecaster was validated by analysing the typical error metrics of sample days from the training and validation databases. The deviation between actual and predicted demand was lower than 0.5% in 97% of the days analysed during the validation phase. Moreover, while the root mean square error was 0.07 MW, the mean absolute error was 0.05 MW. The results suggest that the forecaster's accuracy is considered sufficient for installation in smart grids or other power systems and for predicting future energy demand at the chosen sites.

The aim of this study was to develop an artificial intelligence-based tool that is able to predict wind power density. Wind power density is volatile in nature, and this creates certain challenges, such as grid controlling problems or obstacles to guaranteeing power generation capacity. In order to ensure the proper control of the traditional network, energy generation and demand must be balanced, yet the variability of wind power density poses difficulties for fulfilling this requirement. This study addresses the complex control in systems based on wind energies by proposing a tool that is able to predict future wind power density in the near future, specifically, the next 10 min, allowing microgrid's control to be optimized. The tool is validated by examining the root mean square error value of the prediction. The deviation between the actual and forecasted wind power density was less than 6% for 81% of the examined days in the validation step, from January 2017 to August 2017. In addition, the obtained average deviation for the same period was 3.75%. After analysing the results, it was determined that the forecaster is accurate enough to be installed in systems that have wind turbines in order to improve their control strategy.

This paper proposes an artificial neural network (ANN) to predict the solar energy generation produced by photovoltaic generators. The intermittent nature of solar power creates two main issues. Firstly, power production and demand have to be balanced to ensure the control of the whole system, and the inherent variability of clean energies makes this difficult. Secondly, energy generation companies need a highly accurate day-ahead or intra-day estimation of the energy to be sold in the electricity pool. For the tool developed in this paper, we address the issue of the complexity of control in systems that are based on solar energies. The tool's ability to predict the parameters that are involved in solar energy production will allow us to estimate the future power production in order to optimise grid control. Our tool uses an ANN which we developed using MATLAB (R) software. The results were validated by analysing the root mean square error of the prediction for days outside the database used for training the ANN. The difference between the actually produced and predicted energy is about 0.5-9%, meaning that the accuracy of our tool is sufficient enough to be installed in systems which have integrated solar generators.

Nowadays urban environments concentrate more than half the world's population, reaching up to 70% on 2050 according to forecasts. This concentration implies that most of future challenges will take place in cities as well as the opportunities coming from their potential solutions. Current technological innovation can provide support in facing one of main challenges society is facing: reducing carbon footprint from our cities. This ambitious transition, steered by the Smart Zero Carbon City (SZCC) concept, needs a flexible characterisation method, which can be adapted to different kinds of cities to evaluate the main features of each city, hence proposing and prioritising most suitable interventions. The aim of this study is focused on the characterisation of cities according to the SZCC concept through a set of indicators: the Smart Zero Carbon City Readiness Level (SZCC Readiness Level), able to analyse key aspects of cities according to SZCC concept (Characteristics of the city; City plans and strategies; Energy; Mobility; Infrastructures and ICT services; Citizen Engagement). This characterisation enlightens the development of SZCC concept in the city, identifying its strengths and weaknesses in order to ease the alternatives' selection towards decarbonisation, being handy at a time for those small and medium-sized municipalities, so common in the European context, which usually hold less resources than big capitals to implement decision-making support diagnoses.

This study reviews the causes of power losses in a DC/DC converter, relates those losses with efficiency profiles, and details different strategies for maximising efficiency when several converters are placed in parallel. The research is focused on electric bus applications, where several converters are usually connected in parallel in order to fulfil the power requirements of the bus models. In a bus, the low-voltage DC link (28V) resembles a microgrid that has some generators (DC/DC converters), an energy storage system (lead-acid batteries) and certain loads. To share the converters' output power, a droop control is proposed. Furthermore, traditional droop control is improved by adding a master-slave control. The proposed master-slave droop control improves efficiency by more than 3% at low power with no additional hardware. Experimental results show how two 5kW converters work in parallel using the proposed control.

This study presents the analysis and results of a technique that allows the conventional series resonant DC/DC converter to work as a step-up and step-down converter. The controller can choose between several novel step-up modes and the traditional step-down mode. These modes are analysed by presenting their DC characteristics. The choice of mode is governed by the converter's main controller. The converter reduces the operation frequency range and maximises efficiency by selecting the appropriate mode for the particular load and voltage conversion ratio case. Experimental results in a 6 kW prototype validate the analysis and the proposed controller. Finally, the presented converter is compared with a step-up fixed mode converter in order to appreciate the advantages.

This paper presents the analysis of a bidirectional series-resonant converter (SRC) for light loads working above or below resonance. In normal operation (heavy load), it is well known that the SRCs have a maximum voltage conversion ratio (m) equal to one. This occurs when the switching frequency (f(S)) equals the resonant frequency (f(0)) and power losses are neglected. This maximum can also be obtained for f(S) < f(0) when the switches are turned ON during a half-resonant period. Thus, zero-current switching can be achieved at both the turn-ON and turn-OFF processes. However, anomalous step-up behavior is found when the converter is attached to light loads. A comprehensive analysis of this phenomenon in the time-domain and in the state plane is presented. Based on the latter, an expression that estimates m and shows the influences of the parameters is obtained. With the aid of this expression, a solution that mitigates that effect is presented and implemented in a prototype, which is part of a bidirectional power chain for supercapacitors. This solution can be implemented for both f(S) > f(0) and f(S) < f(0).

This paper proposes a genetic algorithm-based method for sizing the energy storage system (ESS) in microgrids. The main goal of the proposed method is to find the energy and power capacities of the storage system that minimizes the operating cost of the microgrid. The energy management strategy (EMS) used in this paper is based on a fuzzy expert system which is responsible for setting the power output of the ESS. The design of the EMS is carried out by means of a genetic algorithm that is used to set the fuzzy rules and membership functions of the expert system. Given that the size of the storage system has a major influence on the energy management strategy, in this paper the EMS and ESS capacities are jointly optimized. In addition, the proposed method uses an aging model to predict the lifetime of the ESS. In this way it is possible to determine the cost associated with energy storage in a more precise manner. The unit commitment problem, which is crucial for the proper operation of the microgrid, has been also considered in the present work. The suggested sizing methodology has been validated in two case studies.