This paper proposes a multi-objective microgrid (MG) management approach aiming at optimizing operating cost, preserving battery health, and ensuring self-sufficiency in islanding conditions. A novel formulation of self-sufficiency in terms of time duration is proposed, and an optimized MG scheduling is obtained by using a multi-objective Genetic Algorithm (GA) and considering two different criteria to select the best trade-off solution on the Pareto front. The developed energy management strategy was tested on a small-scale residential DC MG. Daily and yearly simulation tests demonstrated the effectiveness of the proposed technique.

This paper deals with current sharing problem for interconnected power converters in DC microgrids. Specifically, it considers as a case study a multi-input converter (MIC), including two voltage sources connected to a common DC bus with a bulk capacitor through two parallel synchronous boost converters and an aggregated load modelled as an ideal current source connected to the DC bus. The dynamics related to current distribution can be controlled without impacting the voltage regulation. This decoupling is performed without resorting to a time-scale separation, which would lower the achievable performance. Instead, unilateral coupling can still be achieved by using a new control structure where the voltage regulation is independent of the current splitting. The proposed strategy, based on dynamic allocation theory, achieves a fast voltage response with an optimal current distribution among the converters, considering the current limits, the dynamic response, and the efficiency of the individual converters. The proposed solution is tested by numerical simulations in a Matlab/Simulink environment.

The increasing integration of batteries in residential microgrids calls for careful consideration of their usage pattern to preserve their state of health (SoH). In fact, using batteries without accounting for degradation phenomena leads to reduced lifetime. This paper proposes a multi-objective Energy Management System (EMS) for a residential microgrid that aims to reduce the user's electricity bill and increase battery life. To suitably account for battery state of health (SoH), a control-oriented physical aging model of the battery has been embedded in the EMS formulation. The choice of such a model allows for effectively incorporating the battery aging process into the algorithm. The optimization problem is solved through dynamic programming, which allows disregarding issues related to model convexity. Different weights between the two objectives and different battery initial ages are used to assess the behavior of the EMS in simulation. Results show an improvement of the battery SoH with the proposed EMS after 100 days in all the scenarios considered.

The use of Energy Management Systems (EMSs) allows obtaining remarkable advantages for both end-users of electrical energy and grid operators. These systems can take advantage of a suitable forecasting of load demand and meteocli-matic variables tied to power generation. In facts, the forecasting ability enables a more effective planning of the power allocation. The aim of this paper is the development of a forecasting module that can be interfaced to EMSs to deliver a 24h ahead forecasting. The module is based on a suitable Artificial Neural Network (ANN), namely the nonlinear autoregressive with exogenous input (NARX) ANN. Such an ANN has been implemented using Tensorflow library and writing Python code. It has been trained using a public solar irradiance dataset, and several tests have been performed to assess its performance with different numbers of output units, hidden layers, and neurons per hidden layer. The obtained results show that the obtained forecasting module has good performance and is suitable for embedded implementation and online operation to support EMSs.

This paper proposes the development of a shipboard Energy Management System (EMS), specifically devised to enhance the efficiency of electrical microgrids in cruise vessels. Due to the target application, the reference electrical plant is an Integrated Power System, typical of modern All Electric Ships (AESs). In order to enable the EMS operation, suitable architecture modifications on the current electrical system have been envisaged, which are also coherent with the ongoing transition of today marine electrical plants toward the microgrid paradigm. In particular, the introduction of energy storage systems (ESSs) has been considered. The proposed EMS performs a real-time scheduling of the electrical power flows, based on load demand forecasting data and a Dynamic Programming (DP)-based optimization strategy. Its main goal is the reduction of the load peaks, which are responsible for unnecessary oversizing of the electrical generators and consequently of inefficient operation. The EMS has been set up using a low-cost programmable embedded platform. The test results on the EMS demonstrate its effectiveness and feasibility.

In order to optimize energy efficiency and to achieve cost savings in smart buildings and grid-connected smart homes that include renewable generators and electrical storage systems, Energy Management Systems (EMSs) are today the most up to date solution. Besides achieving these two goals, a suitable design of the EMS can provide a quite deterministic management of power flows, reducing the gap between actual and predicted power due to forecasting errors. On the basis of a previously proposed EMS that allows reducing both the end-user's electricity bill and the generation/demand uncertainty impact, this paper proposes a detailed analysis of several factors affecting the EMS's performance. Variations of algorithm strategy parameters, market constraints and size of hardware components have been investigated and the results have been evaluated in terms of reduction of power gap and cash flow. Simulation results obtained in a six-day period for a grid connected smart home with a 3 kWp photovoltaic generator and a battery storage system are presented and some guidelines for proper EMS design have been proposed.

The impact of EMI filters on volume and weight of power converters is significant. For this reason, filter's size optimization is a strategic step towards the improvement of the power converter's power density. An EMI filter design that follows a conventional procedure does not guarantee the selection of components/configuration leading to the best power density. Therefore, in order to help EMI engineers and scientists in pursuing a fast and effective choice of optimal discrete EMI filter components and configuration, a novel tool is proposed in this paper, namely ODEF (Optimized Design of EMI Filters). ODEF is an interactive software application running in Matlab®environment. It suitably improves a previously validated EMI filter design procedure that extends the conventional filter design method in order to achieve optimal power density. Features and operation of ODEF tool are illustrated. Moreover, the experimental assessment of an input EMI filter, designed according to the optimized procedure for an inverter-fed induction motor drive, is performed.

This paper presents design issues and realization of a common mode (CM) electromagnetic interference (EMI) filter for a DC-supplied motor drive equipped with an output active CM voltage compensator. The obtained system allows both reliability and EMI of the motor drive to be improved at the same time. In particular, as for reliability, the active CM voltage compensator gives a reduction of the stress on motor bearings; in addition, the input EMI filter, designed taking into account the impedance mismatching between EMI source and receiver in the actual circuit configuration, allows standard limits to be satisfied. Simulation analysis and experimental assessments are given.

In this paper a photovoltaic (PV) battery charger based on a DC-DC boost converter for a small size marine autonomous vehicle (AUV) is developed. The proposed solution employs a neural-based technique to estimate the solar irradiance on the basis of the actual PV panel voltage and current. This information is then used to perform an effective maximum power point tracking (MPPT) to optimise the energy exploitation of the solar panel. In particular the growing Neural Gas Network is used. The design and set up of the PV charger is presented together with experimental results assessing its performance.