A digital relaying algorithm for integrated power system protection and control

Abstract

Recent developments in data packets based high speed digital communications have opened the door for new types of applications in power system protection and control. Intelligent Electronic Devices (IEDs) are equipped with various communication capabilities that make their functional integration a natural next step. Existing integration of substation equipment is not capable of clustering with the purpose of pooling hardware resources. Presently, every electric device requiring protection has its dedicated hardware performing the predetermined set of protective functions. A new function-based protection and control philosophy is proposed, based on an open-system solution. In the proposed system, the resources of the protective and control hardware are pooled, and as a clustered system provide each protected unit (line, transformer, breaker, etc) with functions required for complete direct and backup protection. The work presented in this thesis identifies the performance requirements of a digital relaying algorithm for processing samples that are sent across Ethernet-based communication channels. The work shows the shortcomings and unstable performance of widely used protective algorithms in accommodating data samples that are out of step from their proper position due to variable time delays of the communications media. A new digital relaying algorithm was developed that is able to extract the amplitude and phase angle of signals from data samples received across Ethernet networks with variable jitter. The performance of the algorithm was tested by using the recovered phasor amplitude and phase angle information in protective solutions. The results show that there is significant flexibility in the algorithm that can be used to facilitate less performant communication channels, or, to take advantage of faster communications channels by reducing the response time of the protective function. The results show that the algorithm works well with variable length data windows, and variable sampling frequencies. Higher sampling rates make communications problems more visible, but the presented algorithm is able to compensate for wide variations in network performance, effectively maintaining sampled signal phase and amplitude information during network performance fluctuations.