Novel metal-oxide arrester monitoring technology based on RFID sensor and mind evolutionary computation
Fangming Deng,Kaiyun Wen,Han Zeng,Zhongxin Xie
DOI: https://doi.org/10.1016/j.epsr.2020.106859
IF: 3.818
2021-03-01
Electric Power Systems Research
Abstract:<p>This paper proposes an online monitoring method of metal-oxide arrester (MOA) based on passive RFID sensor tag and mind evolutionary computation (MEC). Firstly, an RFID sensor tag is designed to collect the leakage current of MOA, so as to realize the rapid fault location and equipment life cycle management. The proposed RFID sensor incorporates the sensor data into its ID information. The proposed power management block adopts a new architecture of a low-voltage DC-DC charge pump after a single-stage rectifier circuit. A method of MOA condition monitoring based on MEC is proposed. According to the measured operating voltage and leakage current, the parameters <em>k</em> and <em>α</em> which can reflect the aging condition of MOA are solved by using MEC's better optimization calculation ability, so as to monitor the MOA condition. In addition, the influence of harmonic voltage on the algorithm is analyzed through MATLAB simulation. The results show that the designed sensor tag can effectively measure the leakage current, and the proposed monitoring algorithm can accurately calculate the relevant parameters reflecting the state of MOA. Moreover, under the influence of harmonic voltage, the maximum errors of parameters <em>k</em> and <em>α</em> are 1.8% and 2.0%. Compared with the existing monitoring technology, it has the advantages of low cost, high precision and rapid fault location, which provides a new method and idea for on-line monitoring of MOA.</p>
engineering, electrical & electronic