非破壊検査
Print ISSN : 0367-5866
56 巻, 11 号
選択された号の論文の2件中1~2を表示しています
  • 長 秀雄, 米津 明生, 渡部 剛, 竹本 幹男, 鈴木 裕晶
    2007 年 56 巻 11 号 p. 582-588
    発行日: 2007年
    公開日: 2007/11/21
    ジャーナル フリー
    This paper discusses the location certainty of corrosion zone on floor plates of a cylindrical storage tank by the source location of AE signals produced by rust fracture. For studying the effect of both the AE monitoring and source location methods on the location certainty, two AE monitoring methods and three location schemes were attempted. We first monitored AEs by resonant type sensors mounted on the terrace of the annular plate and on the side wall of the cylindrical tank with naphtha. Source location of the signal AEs selected by waveform classification were estimated by three location schemes using the group velocity of 2300 m/s to study the location certainty of the corrosion zone. Here the certainty of the estimated location was evaluated by the distance error of the locations estimated by three location schemes. Three zones on the floor plate and two zones on the annular plates were estimated as the zones of AE sources. Among them, one zone on the floor plates and two on the annular plates agreed with the zone with large wall reduction.
  • 小山 潔, 星川 洋, 小松 慶亮
    2007 年 56 巻 11 号 p. 589-595
    発行日: 2007年
    公開日: 2007/11/21
    ジャーナル フリー
    The authors propose an eddy current probe with multiple detecting coils for the detection of long slit flaws along the direction of the production line. The proposed probe consists of one rectangular long exciting coil and multiple tangential detecting coils, which are arranged perpendicular to the long side of the exciting coil. The testing of a metal product surface can be conducted by placing the probe so that the detecting coils are parallel to the direction of the production line. It has been confirmed that the proposed probe can reliably detect long slit flaws along the direction of the production line if a group of multiple detecting coils are used. The flaw signal phase varies according to the flaw depth. Thus, the flaw depth can be evaluated by a neural network using the flaw signal phase obtained by multiple detecting coils.
feedback
Top