Self‐adaptive learning based controller to mitigate PQ issues in internet of things devices. (29th March 2021)

• Record Type: Journal Article
• Authors: Goswami, Garima; Goswami, Pankaj Kumar
• Is Part Of: International transactions on electrical energy systems. Volume 31:Number 5(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-03-29
• Subjects: ANFIS -- FLC -- IoT -- NN -- PQ -- THD; Electric power -- Periodicals; Electric power systems -- Periodicals; Electrical engineering -- Periodicals; 621.3
• DOI: 10.1002/2050-7038.12888
• Languages: English
• ISSNs: 2050-7038

Abstract

Summary: The Internet of Things (IoT) systems elicit high speed, precise control and remote access to devices in public and industrial applications. The ease of remote access revokes the issue of system ambiguities due to the presence of hybrid load characteristics. The high‐performance power electronics‐based converters and non‐linear loads pulled out several undesirable power quality issues. Among various performance pursuits, harmonic distortion analysis is at high priority in smart electronic devices. This work analyses the attributes of hybrid load system in smart IoT based devices to observe power quality issues and successful implementation of a smart self‐adaptive learning technique to minimize the harmonic distortions. The model explains the risk analysis and identifies the architectural complexities of the Power Management Unit (PMU) in IoT devices. Hysteresis current control is applied for the generation of reference current to actuate the Shunt Active Power Filter (SAPF) for mitigation of harmonics distortions in distribution supply. Based on real time non‐linear load analysis parameters, ANFIS is trained to compensate the harmonic distortion. The experimental setup validates the error deviation data set for the neural network training and the adaptive control strategy. The analysed empirical weight of the neuron is adopted through various learning layers and minimizes the total harmonic distortion (THD) to a remarkable value from 72.8% to 0.81% at non‐linear load. This validates the high feasibility of the control strategy in various real time smart IoT applications in accordance of IEEE 519 standards. Abstract : The high‐performance power electronics‐based converters and nonlinear loads pulled out several undesirable power quality issues. This work identifies the architectural complexities of Power Management Unit (PMU) in IoT devices and implements a smart self‐adaptive learning technique to minimize the harmonic distortions.