US 12,292,467 B2
Method for identifying and localizing faults in a medium and low voltage electric power distribution grid using measurements from low voltage parts of the grid
Omid Alizadeh-Mousavi, Jongny (CH); and Sotirios Dimitrakopoulos, Zurich (CH)
Assigned to Kraken Technologies Limited, (GB)
Appl. No. 18/552,516
Filed by KRAKENFLEX LIMITED, London (GB)
PCT Filed Mar. 4, 2022, PCT No. PCT/IB2022/051910
§ 371(c)(1), (2) Date Sep. 26, 2023,
PCT Pub. No. WO2022/208184, PCT Pub. Date Oct. 6, 2022.
Claims priority of application No. 21166363 (EP), filed on Mar. 31, 2021.
Prior Publication US 2024/0168075 A1, May 23, 2024
Int. Cl. G01R 31/08 (2020.01)
CPC G01R 31/086 (2013.01) 18 Claims
OG exemplary drawing
 
1. A method for detecting and localizing faults in a medium voltage (MV) and low voltage (LV) distribution grid comprising a plurality of MV and LV nodes and a plurality of MV and LV branches, a plurality of MV/LV transformers each arranged to connect one MV node with one LV node; a monitoring infrastructure comprising metering devices assigned to particular LV nodes of the distribution grid, each of one of said metering devices being arranged to measure a nodal voltage of the LV node to which it is assigned and a current flowing into or out of said LV node through an upstream branch incident on said LV node, each one of said particular LV nodes being one of the nodes that a MV/LV transformer is arranged to connect with a MV node, the monitoring infrastructure further comprising a central processing unit and a communication means arranged for allowing communication between the metering devices and the central processing unit (10); wherein the method comprises the following steps:
I. measuring by the metering device at each one of said particular LV nodes, the voltage v(t) and the current i(t) with a sampling rate that is high enough for sampling said voltage and said current over a bandwidth with an upper limit of at least 1 kHz;
II. calculating by the metering device at each one of said particular LV nodes using the voltages and the currents measured in step I, for each one of a succession of evaluation time windows, the mean value of the nodal negative-sequence voltage (Vneg), and calculating a variation of the calculated mean value with respect to the previous evaluation time window;
III. assessing by the metering device at each one of said particular LV nodes, the following first criterion for each one of the evaluation time windows except the first one: “the variation of the mean value of the nodal negative-sequence voltage (Vneg-rltv) calculated in step II is larger than a first given threshold”, and when the assessment comes out positive for a particular evaluation time window, sending by the metering device a fault notification comprising the calculated variation for the particular evaluation time window to the central processing unit;
IV. identifying by the central processing unit using fault notifications received during a predefined time lag window from metering devices assigned to different LV nodes, the fault location with the help of grid topology information.