In this paper, the discrete wavelet transform (DWT) and a chaotic system were combined with a convolutional neural network (CNN) and applied to the diagnosis of insulation faults in XLPE (cross-linked polyacetylene) power cables. First, four different types of insulation faults in power cables were constructed, including the normal state of the cable, the short outer semi-conducting layer, impurities in the insulation layer, and insulation layer damage, and a high-speed capture card (NI PXI-5105) was adopted to measure the partial discharge (PD) signal, which was then filtered through discrete wavelet transform. Then, based on the Lorenz chaotic system, a dynamic error scatter diagram was established as the feature of each fault state. Finally, the dynamic error scatter diagram was processed by CNN to recognize four different types of faults in the power cable. The test results show that the method proposed in this paper can quickly recognize the fault state of power cables and has excellent performance in terms of recognition accuracy, which reaches 97.5%. Therefore, the proposed method can effectively detect the fault signal changes of power cables and identify the fault state of power cables in real time.
By consulting various worldwide definitions of microgrids and distributed energy, this study presents a microgrid-structured multi-agent system and uses Matlab/Simulink to construct a circuit with microgrid features, which enables the changes in each electrical source and load in the microgrid to be monitored and controlled. This multi-agent system adheres to the Java Agent Development Framework (JADE) platform specifications of the Foundation for Intelligent Physical Agents (FIPA), facilitating communication, information transfers, and the receipt of real-time information regarding the microgrid and each component in the microgrid. Furthermore, the real-time state in the microgrid can be correspondingly controlled, achieving the most efficient real-time monitoring and control for electrical sources and load management in the microgrid.
This research explored the independent solar power system (ISPS), including the maximum power point tracking (MPPT) technology and the DC (direct current) to AC (alternating current) full-bridge inverter (FBI) control technology. This study combined the perturb and observe (P&O) algorithm based on the output voltage control strategy and the proposed optimized FBI control strategy to the ISPS. The proposed variable frequency and variable duty cycle (VFVDC) optimized FBI control strategy had adjustable frequency and duty cycle. The proposed VFVDC control strategy was compared with the sinusoidal pulse width modulation (SPWM). When the output of FBI was at a low voltage level, the operating frequency was 10 kHz. Conversely, when the FBI gave a high voltage level output, the operating frequency was 1 kHz. This control improved the switching loss of 4 power MOSFETs in the FBI and thus improved the performance with a low harmonic amount. To further verify the explored system, this study used MATLAB to compare the proposed control strategy and SPWM control method. The simulation results verified that the proposed control strategy total harmonic distortion (THD) = 1.02% was better than the SPWM control method THD = 3.23%, then the low pass filter (LPF) volume was reduced. Finally, the actual ISPS was used for actual measurement and verification. The proposed control strategy had a low harmonic level, can operate stably and effectively, and thus be used for the actual fulfillment of electricity needs.
INDEX TERMSVariable frequency and variable duty cycle control strategy, maximum power point tracking, solar power systems, full bridge inverter.
This study combined a Convolutional Neural Network (CNN) with the chaos theory and the Empirical Mode Decomposition (EMD) method for the attenuation fault recognition of power capacitors. First, it built six capacitor analysis models, including normal capacitors, failed capacitors, and normal capacitors attenuated by 20-80%. Then a power testing machine was used for an applied voltage test of the capacitor. The EMD method was combined with the chaos synchronisation detection method to chart the discharge signals of the voltage and current that was captured by a high frequency oscilloscope into a 3D chaotic error scatter plot, as the fault diagnosis feature image. Finally, the CNN algorithm was used for the capacitor fault detection. The advantages of the proposed method are that big data are compressed to extract meaningful feature images, the operating state of the power capacitor can be detected effectively, and faults can be diagnosed according to the electrical signal change of the power capacitor. The actual measurement results showed that the accuracy of the proposed method was as high as 97% and has a high efficiency of noise rejection ability, which indicates that the method could be applied to other power-related fields in the future.
Taiwan's power system is isolated and not supported by other interconnected systems. Consequently, the system frequency immediately reflects changes in the system loads. Pumped storage units are crucial for controlling power frequency. These units provide main or auxiliary capacities, reducing the allocation of frequency-regulating reserve (FRR) and further reducing generation costs in system operations. Taiwan's Longmen Nuclear Power Plant is set to be converted for commercial operations, which will significantly alter the spinning reserves in the power system. Thus, this study proposes a safe and economic pumped storage unit dispatch strategy. This strategy is used to determine the optimal FRR capacity and 1-min recovery frequency in a generator failure occurrence at the Longmen Power Plant. In addition, this study considered transmission capacity constraints and conducted power flow analysis of the power systems in Northern, Central, and Southern Taiwan. The results indicated that, in the event of a failure at Longmen Power Plant, the proposed strategy can not only recover the system frequency to an acceptable range to prevent underfrequency load-shedding, but can also mitigate transmission line overloading.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.