MSc thesis project proposal

Estimation algorithm for a Phasor Measurement Unit

Nowadays more and more renewable energy sources are implemented into existing power system networks, which were primarily designed for one way energy flows from power generation facilities to utilization side. Renewables give rise to new conditions and states, to which traditionally designed and operated power systems are not fully prepared. System operation challenges like demand - resource balancing, system reliability, economics, environmental prospective has become more evident and at the same time more challenging to the system operators due to the lack of the advanced systems available for high speed Real Time monitoring and control.

In the mid-1980s synchronised Phasor Measurement Unit (PMU) was introduced and with integration of Internet Technologies (IT), a Synchronized Measurement Technology (SMT) was born. SMT is based on time synchronized high Real Time high speed obtained data, which are essential for Real Time Monitoring and Control of future Power Networks.

PMU serve as a power system voltage and current phasor, frequency and rate of change of frequency (ROCOF) measurement device, capable of sending the synchronous data in correspondence with GPS pps signal. IEEE Standard C37.118.1 (2011) and its Amendment (2014) define PMU output and accuracy requirements for both static and dynamic conditions.

One of the main operations of PMU is high frequency sampling of voltage and current analogue input signals and then calculation of its time synchronized phasors. The voltage signal is of the form

x(t) = A(t) cos(2 pi f_0(1+eps1) t +eps2) + eps3

Assignment

Given the model of the voltage signal, find optimal or efficient estimation algorithms. The implementation may be recursive or adaptive (RLS, Kalman).

The algorithm has to comply to the latest requirements of the IEEE Standard C37.118.1a - 2014, which defines the PMU output and accuracy requirements for both static and dynamic conditions. The algorithm has to be modeled in Matlab/Simulink and compared to other already developed solutions. From the novel developed solution it is expected to be relative fast and computational lightweight, for a possible later FPGA implementation.

Requirements

The student should have followed the Estimation and Detection course. Some Matlab experience is required.

This thesis is in collaboration with the Electrical Sustainable Energy department (dr. Marjan Popov) and in context of an NWO project on Large Power Systems.

Contact

prof.dr.ir. Alle-Jan van der Veen

Signal Processing Systems Group

Department of Microelectronics

Last modified: 2018-03-14