Optimal Parameter Identification of Energy Harvesters Using Vector Impedance Quantum Genetic Algorithm: A Case Study With Current Transformer

• Published in: International journal of energy research 2024-01, Vol.2024 (1)
• Main Authors: Xiang, Shiyezi, Du, Lin, Yu, Huizong, Xiao, Jianhong, Chen, Weigen, Wan, Fu
• Format: Article
• Language: English
• Subjects: Accuracy; Algorithms; Energy; Energy harvesting; Energy technology; Equivalent circuits; Genetic algorithms; Identification methods; Impedance; Load resistance; Methods; Optimization; Optimization algorithms; Parameter estimation; Parameter identification; Performance evaluation; Power supply; Transformers
• ISSN: 0363-907X; 1099-114X
• DOI: 10.1155/2024/6682333

Energy harvesting technology, as an emerging energy technology, has contributed outstandingly to the application of online monitoring sensors in power systems. The accurate parameter identification of the harvesters is crucial for their power supply applications. This work aims to investigate the parameter identification of harvesters with complex equivalent circuit models, here taking a current transformer as an example. However, previous studies focused on optimizing algorithms rather than data sources. This paper demonstrates a vector impedance quantum genetic algorithm (QGA) parameter identification method to identify the amplitude and phase information of the impedance responses. By comparing the results of the proposed method with the genetic algorithm and PSO based on impedance responses and QGA based on load resistance responses, it is proved that the proposed vector impedance QGA identification method is optimal in terms of accuracy, speed, and robustness. The root mean square errors of the output voltage and the phase difference with the primary current for the proposed method are 5.884 × 10 −5 V and 4.473 × 10 −4 ms. Moreover, the practical applicability of this method is validated, demonstrating its effectiveness in real‐life and industrial settings. The proposed identification method in this paper changes the source data so that the sample data are small in volume and extensive in information, which enables faster and more accurate parameter identification. This study provides a new idea for parameter identification researches.