M. Majidi

A Novel Method for Single and Simultaneous Fault Location in Distribution Networks

This paper introduces a novel method for single and simultaneous fault location in distribution networks by means of a sparse representation (SR) vector, Fuzzy-clustering, and machine-learning. The method requires few smart meters along the primary feeders to measure the pre- and during-fault voltages. The voltage sag values for the measured buses produce a vector whose dimension is less than the number of buses in the system. By concatenating the corresponding rows of the bus impedance matrix, an underdetermined set of equation is formed and is used to recover the fault current vector. Since the current vector ideally contains few nonzero values corresponding to fault currents at the faulted points, it is a sparse vector which can be determined by l1-norm minimization. Because the number of nonzero values in the estimated current vector often exceeds the number of fault points, we analyze the nonzero values by Fuzzy-c mean to estimate four possible faults. Furthermore, the nonzero values are processed by a new machine learning method based on the k-nearest neighborhood technique to estimate a single fault location. The performance of our algorithms is validated by their implementation on a real distribution network with noisy and noise-free measurement.

Partial discharge pattern recognition via sparse representation and ANN

In this study, seventeen samples were created for classifying internal, surface, and corona partial discharges (PDs) in a high voltage lab. Next, PDs were measured experimentally to provide a dictionary comprising the types. Due to the huge size of the recorded dataset, a new and straightforward preprocessing method based on signal norms was used to extract the appropriate features of various samples. The new sparse representation classifier (SRC) was computed using ℓ1 and stable ℓ1-norm minimization by means of Primal-Dual Interior Point (PDIP) and Basis Pursuit De-noise (BPDN) algorithms, respectively. The pattern recognition was also performed with an artificial neural network (ANN) and compared with the sparse method. It is shown that both methods have comparable performance if training process, tuning options, and other tasks for finding the best result from ANN are not taken into account. Even with this assumption, it is shown that SRC still performs better than ANN in some cases. In addition, the SRC technique presented in this paper converges to a fixed result, while the results after training the ANN vary with every run due to random initial weights.

Fault Location in Distribution Networks by Compressive Sensing

This paper proposes a novel method for fault location in distribution networks using compressive sensing. During fault and prefault voltages are measured by smart meters along the feeders. The voltage sag vector and impedance matrix produce a current vector that is sparse enough with one nonzero element. This element corresponds to the bus at which a fault occurs. Due to the limited number of smart meters installed at primary feeders, our system equation is underdetermined. Therefore, the l1-norm minimization method is used to calculate the current vector. Primal-dual interior point (PDIP) and the log barrier algorithm (LBA) are utilized to solve the optimization problem with and without measurement noises, respectively. Our proposed method is implemented on a real 13.8-kV, 134-bus distribution network when single-phase, three-phase, double-phase, and double-phase-to-ground short circuits occur. Simulation results show the robustness of the proposed method in noisy environments and satisfactory performance for various faults with different resistances.

A Sparse-Data-Driven Approach for Fault Location in Transmission Networks

This paper proposes an efficient wide-area fault location method to find single- and double-fault locations in transmission networks. Unlike earlier conventional fault location methods, we only require capturing the phasors in a limited number of buses by phasor measurement units (PMUs). The measured voltage phasors and impedance matrix of the system yield an underdetermined system of equations that can be solved by sparse representation recovery methods. The solution gives a sparse fault current vector whose nonzero elements assign the probable faulty zones. For areas with no faults, current phasors measured by PMUs are sufficiently accurate to calculate the adjacent bus voltages. The new calculated voltages and the measured data are used to estimate the faulted lines accurately. The substitution theorem and least-squares method are used to calculate the differences between pre- and during-fault voltages, and currents in both ends of the faulted lines. Transmission line equations developed based on distributed line parameters are then used to pinpoint the fault location along each faulted line. We demonstrate the satisfactory performance of our noniterative methodology and its low computational load using simulations of the IEEE 39-bus test system with noisy measurements, multiple combinations of all fault types, and different resistances.

Compressive Sensing for Power System Data Analysis

Chapter Overview Within this chapter, we will introduce the applications of a state-of-the-art theorem in signal processing and system identification, named as compressive sensing-sparse recovery (CS-SR), in smart power networks monitoring, data analysis, security, and reliability. The sparse nature of the electrical power grids as well as electrical signals is exploited to introduce alternative mathematical formulations to address some of the most famous system modeling problems in power engineering through a compressive signal processing or a sparse system identification framework. First, a short background on CS-SR theorems and techniques is presented. Next, the state of the art in CS-SR applications in smart grid technology is discussed, and finally, the following three data analyses and power network control problems are specifically addressed in detail. The CS-SR techniques are exploited to propose novel methods for distribution system state estimation (DSSE), single and simultaneous fault location in smart distribution and transmission networks, and partial discharge pattern recognition.

Steady-state operation and control of an in-conduit hydro-powered generator

Pumping stations are used to increase water pressures in supply mains for fire fighting, high rise building, and to maintain water supply in water towers and storage tanks. Water pressure reducing valves (PRV) are needed on domestic systems where the municipal water mains pressure is more than 80psi. Some of the energy loss across the PRV can be converted to electricity by replacing the existing PRV with small hydroelectric turbines and generators. Decreasing the water pressure from 150 psi to 50 psi also reduces the water flows through the system and wastewaters. There is a need to develop a scheme to control the operation of the in-conduit hydro-powered generator for maximum efficiency which is the subject of this paper.