Hengxu Ha

Fault detection in Multi-Terminal Modular Multilevel Converter (MMC) based High Voltage DC (HVDC) transmission system

A Multi-Terminal High Voltage Direct Current (MT-HVDC) network is being considered for utilising the full potential of offshore wind power whereas its realisation is currently being hampered by protection issues. In this paper, a protection strategy for future DC grids based on Modular Multilevel Converter (MMC) based HVDC system is presented. Firstly, a fault detection technique based on initial di/dt measurement is presented and thereafter protection strategies for future DC grids are presented. The fault detection technique presented is based on estimating the initial rate of rise of the current, IRRC (di/dt) at fault inception using measured data and thereafter calculating the line inductance. The calculated line inductance is compared with a setting value to determine whether or not a fault has occurred, thus paving the way for a distance protection strategy. Simulations were carried out using Matlab/SIMULINK for varying fault distances. The results obtained show the validity of the technique in detecting and locating DC side short circuits. An advantage of this technique is that it relies only on information from the local end terminal and as such, no communication channel is required, hence satisfying the protection requirement of fast fault detection and location technique for MT-HVDC systems.

Predicting the prospective fault level on distribution grids and its impact on protective relaying

With the integration of various distributed generators (DGs), traditional distribution networks have been transforming into active distribution networks (ADN). One typical challenge of the integration of DGs is that the fault current can possibly exceed the opening capacity of the related breakers. This paper presents a novel distributed solution, which can be hosted in the distributed feeder relays for on-line calculating the predictive fault level only by means of locally measured the voltage and current. Firstly, the equivalent impedance and the equivalent source voltage are calculated by the variation of the measured voltage and current. Then the predictive fault level can be obtained by the calculated equivalent impedance and source voltage of back side. Simulation results show the effectiveness of the proposed altorithm.