Optical Characteristics and Fault Diagnosis of Partial Discharge in C<sub>4</sub>F<sub>7</sub>N/CO<sub>2</sub> Gas Mixture and SF<sub>6</sub> Based on Novel Multispectral Microarray Detection
Yiming Zang, Yong Qian, Xiaoli Zhou, Mohamad Ghaffarian Niasar, Gehao Sheng, Xiuchen Jiang
Abstract
Optical partial discharge (PD) detection is an efficient means of diagnosing the insulation status of power equipment. C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">7</sub> N/CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> gas mixture is a very potential environmentally-friendly SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> substitute gas, and its PD optical characteristics need to be studied to guide the PD diagnosis of novel C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">7</sub> N/CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> equipment. Therefore, this article proposes a multispectral microarray detection technology, which can achieve high-sensitivity detection and PD diagnosis by simultaneously collecting the spectral characteristics of multiple bands. By setting up an experimental platform, the PD experiments of four typical defects in the C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">7</sub> N/CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> gas mixture with five different proportions and pure SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> are carried out. Based on the analysis of PD multispectral features, the correlation between different gases and the difference between different defects are obtained. Finally, by combining multispectral detection with a <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${t}$ </tex-math></inline-formula> -distributed stochastic neighbor embedding (T-SNE) feature extraction algorithm, a PD diagnosis method that can adapt to both C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">7</sub> N/CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> gas mixture and SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> is proposed, which provides a reference for the PD detection of novel C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">7</sub> N/CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> equipment application.