Search

In this article, a robust functional expanded multikernel broad learning system (RFEMBLS) is proposed to compute the complex nonlinear solar photovoltaic (PV) reference voltage more accurately by importing the irradiance and temperature in different uncertainty conditions. The novel droop control mechanism is introduced to obtain reference current to reduce dependence on different connected renewable energy resources. An adaptive integral backstepping sliding mode controller (AIBSMC) is designed to control the dc bus voltage under different abnormal scenarios for the proposed dc microgrid. An asymptotical stability analysis is developed using Lyapunov theory for the PV-battery dc microgrid. A new adoption rule is proposed for the estimation of both references of PV voltage and battery current. Furthermore, the system steady-state error and tracking convergence of the error are improved by adding an integral action. The backstepping method based on the dc bus voltage feedback results in a faster response and negligible chattering. A sliding mode controller is proposed to improve the control precision and robustness. Finally, the proposed RFEMBLS-AIBSMC method is tested using dSPACE platform in the scale-down lab environment to verify the robustness, practicability, feasibility, and efficacy of the proposed system in the real-time scenario.

With the increasing challenge of delivering quality power to the customers along with maintaining voltage stability in the power system, it is desirable that analysis, detection, compression, characterization, and classification can be performed automatically. Further, figuring out the legitimate cause of a power quality disturbance (PQD) is of paramount importance. Therefore, this paper presents a novel multiscale morphological filter-based deep long short-term memory network (DLSTM) with minimum variance random vector functional link network (MVRVFLN) classifier to diagnose and classify various voltage dip scenarios based on the underlying causes. The novel weighted multiscale averaged triple cascaded combination filter (MATCCF) is a fascinated de-noising technique that is designed using four new different cascaded combinations of the basic morphological operators with optimal scale selection based on ensemble kurtosis (EK) index. With direct input of morphological signals, DLSTM results in automatic computation, data compression, feature extraction, and obviation of gradient dispersion issue. Furthermore, the proposed MVRVFLN classifier shows par classification performance with minimum cross-entropy loss and maximum accuracy in noisy and noise-free environments. Finally, the superiority of this novel MM-DLSTM-MVRVFLN methodology (with 99.92% classification accuracy) is validated by comparing it with several deep neural networks (DNNs).

The improved particle swarm optimization algorithm is integrated with variational mode decomposition (VMD) to extract the efficient band-limited intrinsic mode function (BLIMF) of the single and combined power quality events (PQEs). The selected BLIMF of the robust VMD (RVMD) and the privileged Fourier magnitude spectrum (FMS) information are fed to the proposed reduced deep convolutional neural network (RDCNN) for the extraction of the most discriminative unsupervised features. The RVMD-FMS-RDCNN method shows minimum feature overlapping compared with RDCNN and RVMD-RDCNN methods. The feature vector is imported to the novel supervised online kernel random vector functional link network (OKRVFLN) for quick and accurate categorization of complex PQEs. The proposed RVMD-FMS-RDCNN-OKRVFLN method produces excellent recognition capability over RDCNN, RVMD-RDCNN, and RVMD-RDCNN-OKRVFLN methods in noise-free and noisy environments. The unique BLIMF selection, clear detection, descriptive feature extraction, higher learning speed, superior classification accuracy, and robust antinoise performances are considerable importance of the proposed RVMD-FMS-RDCNN-OKRVFLN method. Finally, the proposed method architecture is developed and implemented in a very-high-speed ML506 Virtex-5 FPGA to text, examine, and validate the feasibility, performances, and practicability for online monitoring of the PQEs. [ABSTRACT FROM AUTHOR]

The main purpose of this paper is to detect the power quality events (PQEs) by empirical wavelet transform (EWT) and classify by error minimised extreme learning machine (EMELM). Empirical wavelet transform (EWT) is used to analyse the non-stationary power quality event signals by multi-resolution analysis (MRA). Here, the disturbance energy index feature vector of different electric power supply signals have been acquired by applying the EWT on all the spectral components and to analyse the overall efficiency of the proposed method on both ideal and noisy environments, three types of PQ event data sets are constructed by accumulating the noise of 25, 35 and 45 dB. Extreme learning machine (ELM) is an advanced and efficient classifier, which is implemented to recognise the single as well as multiple PQ fault classes. Based on very high performance under ideal and noisy environment, the new EWT-EMELM method can be implemented in real electrical power systems. The feasibility of proposed method is tested by simulation to verify its cogency.

This paper presents fault detection, classification, and location for a PV-Wind-based DC ring microgrid in the MATLAB/SIMULINK platform. Initially, DC fault signals are collected from local measurements to examine the outcomes of the proposed system. Accurate detection is carried out for all faults, (i.e., cable and arc faults) under two cases of fault resistance and distance variation, with the assistance of primary and secondary detection techniques, i.e. second-order differential current derivative d2I3dt2and sliding mode window-based Pearson’s correlation coefficient. For fault classification a novel approach using modified multifractal detrended fluctuation analysis (M-MFDFA) is presented. The advantage of this method is its ability to estimate the local trends of any order polynomial function with the help of polynomial and trigonometric functions. It also doesn’t require any signal processing algorithm for decomposition resulting and this results in a reduction of computational burden. The detected fault signals are directly passed through the M-MFDFA classifier for fault type classification. To enhance the performance of the proposed classifier, statistical data is obtained from the M-MFDFA feature vectors, and the obtained data is plotted in 2-D and 3-D scatter plots for better visualization. Accurate fault distance estimation is carried out for all types of faults in the DC ring bus microgrid with the assistance of recursive least squares with a forgetting factor (FF-RLS). To verify the performance and superiority of the proposed classifier, it is compared with existing classifiers in terms of features, classification accuracy (CA), and relative computational time (RCT).

[Display omitted]