Correlation analysis between formation process of SF₆ decomposed components and partial discharge qualities

Abstract

To obtain relationship between concentration of SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> decomposed characteristic components (or their concentration ratio) during the formation process and partial discharge qualities and pulse repetition rate, this paper used needle-plate electrode to simulate outthrust insulation fault and conducted a serial of experiments. First, the SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> decomposition characteristic under different PD strengths is obtained. Then this paper defines average discharge qualities per second as Q <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SEC</sub> . It considers both apparent discharge qualities and discharge pulse repetition rate, and analyzes the relation feature between SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> decomposition characteristic (both concentration of component and their effective formation rates RRMS) and Q <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SEC</sub> . Besides, it proposes C(SO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> )/C(SOF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> ) as the energy characteristic ratio (ER) to represent PD energy, and defines the effective energy characteristic ratio as (ER <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">RMS</sub> ). Moreover, it achieves the inner correlation feature between Q <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SEC</sub> and all the characteristic component concentration, the RRMS of each characteristic component and ERRMS. The results show that: since number of C atoms from stainless steel surface and content of trace H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O and O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> in gas chamber were limited, both decomposed component concentration and component concentration ratio accords with the famous "Logistic Population Model" growth trend in ecology, which means they present "S"-curve increase with Q <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SEC</sub> . The proposed ER can effectively indicate the amount of PD energy and the defined ER <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">RMS</sub> can objectively reveal the variance regularity of PD energy with Q <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SEC</sub> . Hence, the correlation mathematic formula between ER <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">RMS</sub> and Q <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SEC</sub> is obtained and ER <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">RMS</sub> can be used as the new characteristic parameter to represent PD energy. It establishes the foundation for the further fault diagnosis research.