Scielo RSS <![CDATA[SAIEE Africa Research Journal]]> http://www.scielo.org.za/rss.php?pid=1991-169620200002&lang=en vol. 111 num. 2 lang. en <![CDATA[SciELO Logo]]> http://www.scielo.org.za/img/en/fbpelogp.gif http://www.scielo.org.za <![CDATA[<b>Selection and Rating of the Step-up Transformer for Renewable Energy Application</b>]]> http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S1991-16962020000200001&lng=en&nrm=iso&tlng=en In the past decade, South Africa has acquired more renewable energy (RE) generation capacity than the rest of the Sub-Saharan countries. This development has led to increasing concerns about problems associated with electrical equipment connected with the RE technologies, especially in wind and solar. These RE technologies have intermittent generation profiles and are connected to non-linear loads. The fundamental electrical equipment including the step-up transformer, with respect to which, high harmonic losses, abnormal temperature rise and gassing problems have been extensively reported within the last 10 years and remains a precedence for many Independent Power Producers (IPPs). A requirement of the step-up transformer is that IPPs are required to provide a technical schedule clarifying the harmonic and distortion content at the plant's point of common coupling (PCC). A lack of this knowledge to the transformer manufactures thereof leads to under-designing or over-designing cases if the harmonic content has been underestimated or overestimated respectively. In hindsight, it may be beneficial for the Independent Power Producers and transformer manufactures to collaborate to assure the transformer design philosophy for the intended RE application is aligned with the technical requirements. In this paper, a method of de-rating the transformer when supplying non-sinusoidal loads is presented. In this method, the continuous power rating of the transformer is reduced to treat the additional losses as a result of harmonic penetration. Initially, a harmonic spectrum supplied by the IPP is used to calculate the transformer load and service losses. Secondly, the harmonic load spectrum is employed to compute de-rating factors ascribed to as "K-Factor" and "Factor-K", indicating the amount of de-rating necessary for the transformer under study when serving the considered harmonic spectrum. Lastly, the thermal considerations under the harmonic spectrum are presented. <![CDATA[<b>Visible Light Communication System Employing Space Time Coded Relay Nodes and Imaging Receivers</b>]]> http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S1991-16962020000200002&lng=en&nrm=iso&tlng=en In this paper, authors propose a VLC system, whose source of information is assumed to be a power line, and employs a combination of relaying and multiple-input multiple-output (MIMO) techniques. Precisely, the performance of a VLC system, with relay nodes equipped with a MIMO technique called space time block coding (STBC), is investigated. The bit error rate (BER) performance of the Alamouti coding (AC) scheme, which is a type of STBC, is specifically considered. In order to enhance the overall system reliability, we propose to use an imaging receiver (ImR), in conjunction with the relay-MIMO setup. In comparison to when a nonimaging receiver (NImR) is employed, an improved signal-to-noise ratio (SNR) of 7 dB is observed when an ImR with a specific number of pixels, is used, hence, improving the BER. We also investigate and show that the selected number of pixels and receiver location have an effect on the system BER. SNR values of 9 dB, 11 dB and 15 dB were required to maintain a BER of 10~4for imaging receivers with 70, 50 and 20 pixels, respectively. <![CDATA[<b>Stochastic Assessment of the Impact of Distributed Synchronous Generators on Voltage Sags due to System-Wide Faults</b>]]> http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S1991-16962020000200003&lng=en&nrm=iso&tlng=en In this paper, the authors use a stochastic approach to investigate the impact of distributed synchronous generation (DSG) on the severity of voltage sags due to system-wide occurring faults. Based on the method of fault positions, faults of various types are applied at pre-determined locations along all lines in the system and remaining voltages on the faulted phase at a monitoring point are recorded. This is done for a base case and a case with DSG. The results show with DSG, there is a general improvement in magnitudes of remaining voltage. Comparison of the populations of voltages for the two cases, using The Wilcoxon Signed Rank Sum Test, showed that these improvement are statistically significant. Further analysis showed that DSG either reduces severity of or eliminates some voltage sags. In addition, the number of expected trips due voltage sags is reduced with DSG, for any voltage sensitivity level assumed. The study concluded that incorporation of distributed synchronous generation is beneficial to the improvement of voltage sags and this should be a further incentive for their integration, especially into weak networks. <![CDATA[<b>Economic impact of electricity supply interruptions in South Africa</b>]]> http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S1991-16962020000200004&lng=en&nrm=iso&tlng=en In past studies on the costs of interruptions of electricity supply in South Africa, the concepts of sporadic and chronic interruptions were introduced. Cost was modelled using different parameters, and the measurement of customer interruption cost (CIC) was acquired through surveys. Rotational load shedding, as experienced during more than a decade in South Africa, shares many characteristics with chronic interruptions and large system collapse scenarios. Since CIC data are based on electricity customers' valuations of their impacts from electricity interruptions, and allow only for a bottom-up estimation of the economic cost of large electricity interruption events (without considering sectoral interdependencies), an alternative assessment of the cost using a suitable macroeconomic model that employs less subjective data allows for a validation of CIC-based results and a determination of a plausible range of estimates of the cost of large electricity interruptions. We test this proposition using a combination of a time-dependent probabilistic CIC model based on CIC survey data, and a dynamic inoperability input-output model (DIIM) that accounts for sectoral interdependencies, economic resilience and the temporal variation of electricity interruption impacts. The results lead to estimates of the costs of large interruptions of electricity supply in South Africa.