Abstract This article presents a sound pressure identification method to detect internal leak of safety relief valve (SRV) during the sealing surface damaged between valve disk and valve seat. An adjustable acoustical generator was designed to generate ultrasonic waves and detect internal leak. This approach was established by using computational fluid dynamics (CFD) with large eddy simulation (LES) to compute flow field and applying ACTRAN software with Lighthill’s acoustic analogy model to simulate acoustic field. The performance of the proposed method is evaluated by testing under the different leakage rates, as 0.5 L/min, 1 L/min to 5.5 L/min with every 0.5 L/min, respectively. Two microphone sensors were installed at external designed added device around the SRV to capture sound pressure level (SPL) of internal leakage signal. In order to ensure safety of the experiment, the high temperature dry saturated steam was replaced by compressed air as flow (10 MPa). The experimental results show that the proposed method can efficiently identify internal leakage rate higher than 1 L/min, and the sensitive frequency ranges from 7 kHz to 40 kHz. Hence, the acoustic signal of leakage can easily distinguish from noise because ambient noise focuses on low frequency domain. The study makes it possible to detect internal leak of SRV and prevent catastrophic failures in nuclear power plants.

中文翻译：

---

核电站安全泄压阀内部泄漏声压识别方法的研制

摘要 本文提出了一种声压识别方法，用于检测安全泄压阀（SRV）阀盘与阀座密封面损坏时的内部泄漏。可调式声波发生器设计用于产生超声波并检测内部泄漏。该方法是通过使用计算流体动力学（CFD）和大涡模拟（LES）来计算流场，并应用带有Lighthill声学类比模型的ACTRAN软件来模拟声场而建立的。通过在不同的泄漏率下进行测试来评估所提出方法的性能，分别为 0.5 L/min、1 L/min 至 5.5 L/min 和每 0.5 L/min。两个麦克风传感器安装在 SRV 周围的外部设计附加设备上，以捕获内部泄漏信号的声压级 (SPL)。为保证实验安全，将高温干燥饱和蒸汽改为压缩空气作为流量（10 MPa）。实验结果表明，该方法能有效识别大于1 L/min的内漏率，敏感频率范围为7 kHz～40 kHz。因此，由于环境噪声集中在低频域，泄漏的声学信号可以很容易地与噪声区分开来。该研究使检测 SRV 的内部泄漏和防止核电厂发生灾难性故障成为可能。因此，由于环境噪声集中在低频域，泄漏的声学信号可以很容易地与噪声区分开来。该研究使检测 SRV 的内部泄漏和防止核电厂发生灾难性故障成为可能。因此，由于环境噪声集中在低频域，泄漏的声学信号可以很容易地与噪声区分开来。该研究使检测 SRV 的内部泄漏和防止核电厂发生灾难性故障成为可能。