Proliferation of distributed generation units, integrated within the distribution network requires increased attention to their proper placements. In urban areas, buildings' rooftops are expected to have greater involvement in the deployment of PV (photovoltaic) systems. This paper proposes a novel procedure for determining roof surfaces suitable for their installation. The PV potential of roof surfaces is assessed based on Light Detection And Ranging (LiDAR) data and pyranometer measurements. Then, the time-dependent PV generation profiles, electricity distribution network configuration, and time dependent loading profiles are used together over time-steps for selecting those roof surfaces with the highest PV potential, which would lead to the highest reduction of network losses per year. The presented procedure was implemented within a real urban area distribution network. The results obtained confirmed that PV potential assessment could be an insufficient criterion when selecting those roof surfaces suitable for the installation of PV systems. In order to obtain relevant results, network configuration and time-dependent loading and generation profiles must be considered as well. (C) 2015 Elsevier Ltd. All rights reserved.
COBISS.SI-ID: 19317526
Within the last few years, the increase of the world's energy consumption has substantially impacted the environment. Solar energy initiative is more than ever involved to tackle this issue, especially when deploying PV (photovoltaic) systems over large-scale residential areas. However, not all surfaces in these areas are economically suitable, while some surfaces have low CO2 mitigation. With the availability of high-resolution remote sensing data, the estimation of suitable rooftops for PV systems installation can be performed automatically by estimating the PV potential. This paper presents a novel method for estimating NPV (net present value) of the potential PV systems installed on rooftops by using LiDAR (Light Detection And Ranging) data and PV systems' nonlinear efficiency characteristics. More importantly, the environmental impact is estimated for each rooftop through EPBT (energy payback time) and GGER (greenhouse gas emission rate), based on the life-cycle of a specific PV system. This is combined with NPV in order to find rooftops that are both economically and environmentally viable candidates for PV systems deployment. Results demonstrate a case study LiDAR data for predicting each building's economical and environmental impact, as well as providing an overall view of resulting cumulative CO2 mitigation over large residential area. (C) 2016 Elsevier Ltd. All rights reserved.
COBISS.SI-ID: 19549718
This paper discusses methods for estimation of area's frequency response characteristic (AFRC) during large frequency changes. The operating point changes significantly during such transient events; consequently the estimation method should adapt to the time-varying nature of the AFRC. Moreover, available measurements have a poor time resolution and are considerably contaminated by errors. Therefore, new estimation methods are proposed that are based on local correlation between the frequency deviation and interchange power variation measured on all area's tie-lines. The presented estimation methods were compared with a classical approach that is based on the ratio of mean changes as well as with conventional least squares parameter estimators. Extensive numerical simulations and field measurements were applied for performance evaluation, where the obtained results show robust and satisfactorily accurate responses of the proposed estimation methods.
COBISS.SI-ID: 19552534
This paper compares two different identification methods of a static rate-independent energy-based hysteresis model with regard to the dynamic hysteresis loop shape prediction when coupled to the parametric magneto-dynamic lamination model. The values of hysteresis model parameters are determined solely based on the quasi-static major loop. A semi-physical approach identifying the reversible and irreversible field components independently and a purely-mathematical using a differential evolution optimization algorithm determining all parameters simultaneously are compared. Both variants of parameter identification are analyzed in terms of hysteresis loop shape prediction for quasi-static as well as dynamic loops.
COBISS.SI-ID: 19526934
This paper presents one-dimensional dynamic magnetization models of non-oriented soft magnetic steel sheets (SMSSs) that can be expressed as simple systems of ordinary differential equations. The discussed models take into account the dynamic effects on magnetization due to eddy currents and hysteresis inside such sheets, and differ in the way the coupled Maxwell equations with hysteresis are solved. The presented modeling approaches include finite-difference schemes of different accuracies, various magnetic equivalent circuits (MECs) including a recent approach to eliminating the deficiencies of classical MECs, and a mesh-free approach. The different modeling approaches are analyzed and compared in terms of mathematical structure, implementation work, spatial discretization and accuracy, where both voltage- and current-driven versions are investigated.
COBISS.SI-ID: 19374614