Investigation of Statistical Distribution of Energization Overvoltages in 380 kV Hybrid OHL-Cable Systems Hossein Khalilnezhad, Marjan Popov, Jorrit A. Bos, Jan P. W. de Jong, Lou van der Sluis Abstract—Switching operations in power systems can produce significant overvoltages under specific circumstances. With the increasing application of underground cables in transmission systems, the statistical distribution of energization overvoltages is expected to change substantially due to the different electrical characteristics of cables and OHLs. Therefore, it is crucial to perform an insulation coordination study by analysis of the statistical distribution of energization overvoltages. This paper presents a statistical switching analysis on a hybrid OHL-Cable circuit to investigate how such hybrid circuits can affect the distribution of overvoltages. The literature has addressed the distribution of energization overvoltages only for OHLs or cables, but such an study is not available for hybrid systems consisting of OHLs and cables combined. The study is carried out for different cable lengths in the case study to identify how an increasing cable share in the circuit influences the overvoltages distribution due to no-load energization. Moreover, the impact of symmetrical and asymmetrical circuit structures is also addressed. The study is carried out on a distributed frequency-dependent parameter model of the Dutch 380 kV grid in PSCAD/EMTDC. Keywords: Cable, energization, insulation coordination, switching overvoltage, statistical switching. I. INTRODUCTION RANSMISSION System Operators are nowadays investigating the use of long stretches of (E)HV underground cables for grid reinforcements and expansions. This is due to strong political and social oppositions against building new overhead lines (OHL). The trend in installation of longer cables leads to increasing challenges from the system technical operation aspect, mainly the system transient behavior [1]. There are significant differences in electrical characteristics of a cable and an OHL. Shunt capacitance of a cable can be up to thirty times larger and series inductance can be up to five times smaller than those of an equivalent OHL [1]. The capacitive characteristic of cables increases the concerns regarding significant switching overvoltages occurrence, especially during cable no-load energization. This research is financially supported by TenneT TSO B.V., Arnhem, the Netherlands, within the framework of 380 kV cable research program. H. Khalilnezhad, M. Popov, and L. van der Sluis are with the department of Electrical Sustainable Energy, Delft University of Technology, Mekelweg 4, 2628 CD, Delft, the Netherlands (e-mail: H.Khalilnezhad@tudelft.nl, M.Popov@tudelft.nl, L.Vandersluis@tudelft.nl). J. A. Bos and J. P. W. de Jong are with the Grid Development and Strategy Department, TenneT TSO B.V., Utrechtseweg 310, Arnhem, the Netherlands (e-mail: Jorrit.Bos@tennet.eu, Jan.P.de.Jong@tennet.eu). Paper submitted to the International Conference on Power Systems Transients (IPST2017) in Seoul, Republic of Korea June 26-29, 2017. The transient overvoltages caused by energization of an unloaded cable are oscillatory with frequencies up to 20 kHz and duration of several milliseconds [2]. The amplitude, frequency, and duration of energization overvoltages are determined by several parameters, mainly the short-circuit strength of the busbar from which the cable is energized, cable length, shunt compensation size, and breaker closing time. The circuit-breaker closing time depends on the angle (point-on-wave) of the power-frequency voltage waveform where the breaker receives the close command and also on the pole closing span of the circuit-breaker. The pole closing span refers to the time difference between the first and the last pole to close (due to the different unpredictable mechanical delays) and it has a statistical behavior according to a normal distribution. Thus, the circuit-breaker closing time has a random behavior and a statistical approach, by taking into account the statistical behavior of switching time, is the most practical way to analyze the distribution of switching overvoltages [3]-[6]. Analysis of the statistical distribution of overvoltages is crucial in insulation coordination studies. Such an analysis helps to determine the optimum insulation level, to minimize the cost, and to increase system reliability by means of finding the most severe overvoltages and planning proper countermeasures when overvoltages exceed withstand voltages of system components. This paper performs a probabilistic analysis on the transient overvoltages occurring due to no-load energization of a hybrid OHL-Cable transmission line. The statistical distribution of energization overvoltages is addressed in literature for OHL and cable individually, but it is not available for hybrid systems consisting of OHL and cable combined. The study comprises time-domain simulations in PSCAD/EMTDC, where an accurate distributed frequency-dependent parameter model of the whole Dutch 380 kV grid is developed. The influence of the cable length on overvoltages distribution is investigated by studying different cable scenarios in a case-study hybrid OHL-Cable project with 80 km transmission length. Mixed-line is the term sometimes used for hybrid OHL-Cable circuits composed of solidly series connected OHL and cable sections. The effect of the mixed-line structure on the overvoltages distribution is also evaluated by modeling symmetrical and asymmetrical structures. There are four key values obtained from a statistical analysis, namely: the maximum value, mean value, standard deviation, and 2% value. These indicators are used to assess the studied scenarios. The simulation results show that the probability of occurrence of high overvoltages decreases by increasing the cable share in the hybrid OHL-Cable circuit. In other words, T