Application Of Mathematical Morphology In Power System Protection

The rapid and continuous development of power system in terms of capacity, size and voltage level has brought higher requirements to protective relay than ever before. It can be seen by reviewing the history of power system protection, novel theories, techniques, material have been providing novel means to advance power system protection. Nowadays, as the result of the development in microprocessor, communication and digital signal processing technology, power system protection has become a part of information technology. Signal processing technology is one of the key factors enable the development in digital relay, which utilizes mathematical methods to estimate, analyze and identify fault information from discrete sampled sequences. Mathematical morphology (MM) is one of the lastest outcomes in advanced signal processing field. The development history of MM is reviewed at the beginning of this dissertation, some fundamental concepts are then introduced. After that, two novel morphological algorithms suitable for power system protection application are put forward. Afterwards, some attempts to use these two algorithms to improve some protection principles or implementations are made.MM is developed from set theory and integral geometry, and is widely used in geometrical analysis and description. It is concerned with the shape of a waveform in time domain rather than frequency. Morphological filters are nonlinear signal transformation that locally modifies geometrical features of signals and image objects. Morphological calculation needs only the comparison and addition operations without multiplication, the computational complexity, therefore, is much smaller than conventional digital filter algorithms.Driven by the purpose of application in protective relaying, two novel morphological algorithms, series multi-resolution morphological gradient (SMMG) transformation and multi-resolution morphological opening-closing (MMOC) filter, are proposed in this dissertation. The former can be used as a singularity detector and the latter can be used to suppress various types of noise interference mixed with sampled signals. The definitions and mechanisms of these algorithms are given, the length of required data-window is then formulated.SMMG filters is proposed to detect the faint second changes of sequence currents and implement accelerated trip of power transmission lines, while a fault occurs in the second zone of distance protection. A self-adaptive accelerated trip scheme based on SMMG transform is presented. The proposed scheme can considerably improve the sensitivity of accelerated trip and the calculation complexity is much smaller compared with conventional algorithms. The EMTP simulation and test on a dynamic power system model verify the feasibility and efficiency of the proposed approach.The speed and accuracy of digital relays of the transmission line can be improved by accurate and fast fault phase selection and this also allows single pole tripping and autoreclosure to be employed. The present phase selection schemes, however, have defects. A fault phase selector based on SMMG is proposed in this regard. The proposed scheme identifies fault phases by using a SMMG filter to extract the power frequency fault components (PFFC) of the modular currents. The computational burden is lighter than that of conventional methods. The approach has a high speed, sensitivity and reliability. The EMTP simulation result verifies its feasibility.In order to prevent the distance protection from tripping during power swing conditions, a power swing blocking device is often utilized. On the other hand, it has been an increasing requirement to achieve rapid clearance of internal faults during power swings. Accurate phase selection is a prerequisite to selective clearance of faults. An improved unbalanced fault phase selector during power swing based on SMMG is proposed. As a feature extractor from raw signals, SMMG is employed to extract superimposed (fault component) current under power swing condition.The differential relay possibly mal-operate due to the inconsistent transforming characteristics of current transformers (CT) when experiencing heavy through fault. According to the investigation of CT of differential protection, the time interval between the occurrence of differential current and the sudden change of phase current can be utilized to distinguish between the external faults and the internal faults. A new CT saturation-blocking scheme for differential protections, which aims at identifying this time interval by virtue of SMMG, is proposed. Using the excellent singularity detecting ability of SMMG, this time interval can be identified in real time with high accuracy.Multi-resolution morphological analysis (MMA) is introduced to protection relay in this dissertation. An approach to Ultra-High-Speed (UHS) direction protection of power transmission line is developed based on MMA. In the proposed scheme, a MMOC filter is used to suppress interference noise mingled with transient traveling wave. After that, a 1-level Multi-resolution Morphological Gradient (MMG) transform is designed to identify the polarities of current and voltage waves, the fault direction is recognized accordingly. The feasibility and efficiency of the proposed UHS directional protection algorithm are verified by EMTDC. According to the simulation results, the proposed MMOC filter can effectively remove intense noise from the transient waves and polarities of the traveling waves can be exactly identified by the MMG transform even under the worst-case scenarios such as close in faults or fault inception angle at zero crossing.All the theoretical innovations and application studies are summarized at the end of this dissertation, where a few prospective issues worthy studying are also proposed.