This study develops a hybrid method to identify, classify, and locate electrical faults on transmission lines based on Machine Learning (ML) methods. Firstly, Wavelet Transform (WT) technique is applied to extract features from the current or voltage signals. The extracted signals are decomposed into eleven coefficients. These coefficients are calculated to the energy level, and the data of teen fault types are converted to the RGB image. Secondly, GoogLeNet model is applied to classify the fault, and Convolutional Neural Network (CNN) method is proposed to locate the fault. The proposed method is tested on the four-bus power system with the 220 kV transmission line via time-domain simulation using Matlab software. The conditions of the fault resistance random values and the pre-fault load changes are considered. The simulation results show that the proposed method has high accuracy and fast processing time, and is a useful tool for analyzing the system stability in the field of electricity.

This paper presents a new method of current control for the cascaded three-level inverter based on fuzzy logic sliding mode control (FLSMC). To minimize the chattering caused by fast and large amplitude changes of currents and voltages across the load, a fuzzy logic observer is utilized to identify the switch control law. The proposed FLSMC method shows that the measured current can closely follow the set current, and the high harmonic components of the output voltage are low. Additionally, during the operation of the multilevel inverter system, the common-mode voltage (CMV) is also reduced. The proposed method can reduce the number of switching times, which results in lower losses due to the switching process of the components. The proposed FLSMC is based on a control algorithm, not relying on carrier waves to generate PWM pulses, unlike the conventional inverters of the past. Instead, it generates PWM pulses by comparing the control signal with voltage levels. The simulation and testing results have demonstrated the superiority and high efficiency of the proposed FLSMC method.

The rapid development of electric vehicles, electric vehicle charging stations, renewable energy harvesting and storage systems, and various other energy conversion systems has increased the focus on multilevel inverters. These inverters are highly effective in reducing harmonic distortion. However, Common-Mode Voltage (CMV) remains a significant challenge, causing negative impacts, such as reduced component lifespan, decreased control system reliability, and substantial electromagnetic interference. This study introduces a straightforward and highly practical technique that ensures effective CMV reduction. The technique enhances the traditional SPWM method by offsetting the control signal to optimize switching frequency and produce better PWM signals. This method, referred to as Optimal PWM Signal Modulation (OSM), has shown superior CMV reduction capabilities for multilevel inverters compared to the phase disposition technique. The effectiveness of the OSM technique has been evaluated through simulations and practical experiments on a three-level H-bridge converter and the LAUNCHXL-F28379D development kit from Texas Instruments.

This paper presents a study of FACTS devices applications, in particularly choice of adequate shunt compensators (type of FACTS devices, location) in order to improve voltage and transient angle stability. The study is carried out for the 500kV Southern Vietnam power system in 2020 with a high load scenario. The load flow calculation, P-V curve analysis and transient stability study are examined by using the PSS/E-33 software. The obtained results show that with the application of FACTS devices, voltage stability and transient angle stability of this power system are considerably improved.

This paper presents a new study of power system transient fault recognition using Wavelet Multi-Resolution Analysis (MRA) technique integrated with Neural Network. The proposed method requires less number of features as compared to conventional approach for the identification. The feature extracted through the wavelet is trained by a Probabilistic Neural Network for the classification of events. After training the neural network, the weight obtained is used to classify the Power Quality (PQ) problems. These techniques are applied to recognize different faults in the supply voltage of the Southern Vietnam power system at NHABE substation. The research results prove the techniques can be used to detect and classify a wide range of power different faults occurring in power systems with a high accurate ratio. The simulation results possess significant improvement over existing methods in signal detection and classification.