No abstract

No abstract

Abstract. The adequate protection of the distribution transformers is a critical issue, in an effort to avoid interruption of the power supply and damages of the equipment. The installation of metal oxide gapless surge arresters at the entrance of the transformers improves their lightning performance and reduces the expected annual failure rate. However, arresters can be damaged, if the discharge current that passes through them is high enough, resulting in thermal stress of the device and possible failure. The parallel connection of arresters contributes to equal current sharing, reducing their failure probability. In the current work, the parallel combination of two metal-oxide gapless surge arresters installed at the MV side of a distribution transformer is examined, considering how the voltage -current characteristic of each arrester influences the expected voltage surges at the entrance of the transformer, the current sharing between the non-linear resistors and the arresters failure probability.Keywords: Substations, lightning, transformer, surge arresters, parallel connection IntroductionThe reliable and uninterrupted operation of the medium voltage/low voltage (MV/LV) substations of the distribution electrical networks is necessary, in order to ensure the continuity and the quality of the power supply. The MV/LV transformer is the most important component of the distribution system that has to be adequately protected against several factors, considering various technoeconomical and safety parameters. Lightning overvoltages are the main cause of damages and outages in the distribution substations [1]. The implementation of overhead ground wires is not recommended in the case of MV installations, considering the low insulation level and the difficulty in improving the grounding resistance and the insulation withstand capability, in an effort to prevent backflashover phenomena [2][3][4]. The installation of metal oxide gapless surge arresters is the common practice to improve the lightning performance of the network and restrain the arising overvoltages that could stress the insulation of the equipment [5][6][7][8]. However, in the case that the energy that passes through the arresters exceeds their withstand capability a failure is occurred. The failed arresters do not protect anymore the substation and have to be replaced, something that is not always immediately attainable, since the operation of the arresters is rarely monitored. So, the transformers remain unprotected, and severe damages can be occurred. The parallel combination of arresters definitely reduces the absorbed energy of the arresters, since the lightning current is shared; furthermore, the residual voltage is also altered, according to the voltage -current characteristic of the non-linear resistor of the arresters [9,10]. In the current work, the impact of the parallel combination of two metal-oxide gapless surge arresters installed at the MV side of a distribution transformer is examined, revealing the influence of the voltage -current ch...

No abstract

According to the literature high voltage transmission line problems are faced using conventional analytical methods, which include in most cases empirical and/or approximating equations. Artificial intelligence and more specifically artificial neural networks (ANN) are addressed in this work, in order to give accurate solutions to high voltage transmission line problems using in the calculations only actual field data. Two different case studies are studied, i.e., the estimation of critical flashover voltage on polluted insulators and the estimation of lightning performance of high voltage transmission lines. ANN models are developed and are tested on operating high voltage transmission lines and polluted insulators, producing very satisfactory results. These two ANN models can be used in electrical engineers' studies aiming at the more effective protection of high voltage equipment.